Cohort 7 - Day 1 - 4-21-2025

And we're gonna get. Okay.

All right, well, welcome guys. Really excited to have you all in. We're gonna do some fun things over the next three days. It's gonna go really, really fast. I know three hours a day seems like a lot, but it's really not. And it's going to take you, you know, a little bit of time to, to kind of, you know, get used to what we're doing. But I, I think within about an hour, you guys are going to start to feel pretty empowered and by the.

End of today you're going to be.

Like, holy cow, like my whole life has already changed.

Usually by the end of the first day, people are really cranking in a big way with everything that's going on.

Welcome, Devin.

Thank you.

And really excited to, really excited to help you guys. Many of you I've spoken with in depth about this.

Some of you are on because your.

Business is going to be doing some AI training as well.

Maybe you are on here so that.

You can learn AI and then you.

Can go and teach your business. And that's awesome too.

But, but the reality is, guys, there's there.

There's a huge, huge, massive change that is happening in the world right now that's happening at a speed that's actually quite frightening. And there will be a moment or two in this course where you're going to be a little bit scared for.

The world as you learn a little.

Bit more about how fast this really is going to.

But the good news is you guys.

Don'T need to be afraid because you're.

Sitting here and you're learning.

You're staying aware and awake and, and really, really ready to. Really ready to rock here.

So with that being said, I want to just really quickly do some, some housekeeping. I want to make sure everybody. So did everybody get the email from.

That was sent out last night? Okay. I know it was Easter, so, you know, I know that I probably should have sent it out Friday or Saturday, but I was like doing yard work and cleaning up my five acre ranch, getting ready for a ton of people for Easter.

So I apologize for that. That's really on me. Normally I send it out a couple days in advance, but there's not a whole lot on there. But the, the one thing I want to make sure you do all have is. And if you don't have it right now, you can do it while we're sitting here in the chorus right now. We're going to dive in pretty fast though. The first thing I want you to make sure. You have up and running is either an X Premium account or a Grok.

Account if you just want to go directly to Grok.

But Grok is an AI just like.

Kind of like Chat GPT. Right?

X Premium is cheaper if you buy.

It on the on desktop instead of through the app.

So if you already bought it through.

The app, you know it is what it is. Cancel it and you know, let it, let it expire and then re sign up on, on the website later. It's like because the reason it's more expensive on the app is because of Apple's surcharges, all those things. So it's cheaper if you just go directly there.

Chad GPT is actually the same price on the app and on the website.

Whatever. So I guess they're just taking the hit or somehow they have something worked out with Apple.

And then. So make sure you have Chat GPT. You're gonna need chatgpt plus for this.

For this, for this class to work in the way it needs to.

And then the third piece of software.

You'Re going to need is replit.

Okay, we're not going to go over replit in the first probably hour today. I might, I might pull it up and show you a little bit, but we're going to start in Grok and we're just going to kind of dive right in and then we'll do some introductions and things like that after we.

After we kind of get going.

So, so what I want you guys to do right now is you're just going to watch me and then we're going to, we're going to all dive. So before we, before we get started, I actually need a quick inventory on this and I'm just going to go in the order that I think it's in the order that you joined. So Trevor, Brian, Travis, Eric, Jessica, Evan, Devin.

Evan, Devin. That's a nice way to finish out there.

But I want to know, I want you to just. We're going to do now, we're not.

Going to do interest right now.

I kind of want to do interest, but I just want to ask this, do you have any coding experience?

I want to know if you know.

How to develop and so we'll go one at a time. Trevor.

Yeah, I'm pretty much a principal engineer.

Perfect. That's what I need to know.

Brian?

Nothing.

Travis.

I only know how to use cursor.

And a little bit of replit.

Awesome. You're already ahead. That's awesome. Eric.

Yep.

0 Jessica 00 Evan 0 okay, and then yeah, yeah, yeah, we got you Devin. 00.

All right, perfect guys, here's the good news. I'm with the zeros on here, all right? I don't know how to code. I'm not a developer. Now I like some of you guys.

Have been the CEO of a tech.

Company and been around code and around developers for you know, the better part of a deck the last decade. And really, you know, I was in.

Tech companies before that for another six or seven years.

So I've been around it. A lot of you guys have been around it. A lot of you probably, you know, employ developers and so you sit in some of those meetings and where you're architecting and trying to figure out, you know, hey, does this database work? Can we scale to this number or what's going to happen here? You know, so you've had some questions like that. Things like that will help you in this course but again are not required. We're going to. If you have zero experience, Jessica, I'm going to pick on you because I know you. Jessica is the best friend of our my cousins and known her my whole.

Life like for 30 years, forever.

I don't know anyway, but I know she has 000 coding experience and that's perfectly fine. But all the way down to Jessica in terms of coding experience, you will be, you'll be just fine.

We'll even do a little about a 10 minute coding boot camp here in.

A little bit at least around terminology. So okay, perfect. That's what I want to know. And then for those of you, for those of you that are really good at developing, you know, I, I love your input and I love the types of things you can bring to the table. But you know, one of the things.

That'S just huge here is.

I raised $13 million for my last company, a.

Company called Seek that I started back.

In 2016 and we, we built software. Right. Evan is my brother in law actually sitting on here as well. He built the company also raised some.

Money and then spent a huge amount.

Of money paying a big team of developers. Right? I mean a lot of, a lot of us have done that and what is now possible today to do with, with code and even if you don't know code, right? But to be able to build software requires, especially at the beginning, entrepreneurship level, right. The title of this course is how.

To build a business with AI in three days.

Now some of you might be taking the course just because you want to learn AI better but we're going to be teaching it in the context of how to build a business. And in doing that, we're going to go over easily 15 different AI platforms and programs.

Some of them just briefly, some of them a little bit, some of them vary in depth.

We're going to learn how to connect them together and have them bounce off each other and how to argue with each other, things like that. Right. So that you can really boil down to really, really solid answers. But at this point, it is possible to build and start a software business to the point where you not only have an mvp, but where you have your first customers, maybe even hundreds or thousands of customers, depending if you're B2C maybe, you know, But I think you.

Can take it up to at least.

Several million in revenue, if not a lot more than that, without spending more.

Than maybe a couple thousand dollars of development work.

This business, this business, Genaipi, that I started in three days has generated about $200,000 of revenue. And I started it on. Well, I had it in my mind for a long time, and I had.

Some kind of ideas and plans and things like that.

But I started writing the first line of code with AI on February 28th, and I've generated over $200,000, completely validated.

The entire business model. And, you know, I've spent very, very.

Little and many of you interacted with my website to sign up for it. Everything that you saw, everything was there.

Everything was done with AI.

It's not perfect yet. It's definitely not perfect yet. Plenty of little bugs and things I need to fix and whatever, but. But it is more than good enough.

To be completely validated as a business model. And I've spent less than $1,000 so far.

I spent roughly $120 or so in.

That weekend building the entire thing out.

And by.

By a few days later from that, I had my first $15,000 contract.

Okay, so we're talking about a lot about entrepreneurship. We're going to be talking about AI and using all these tools, and we're going to be learning from each other. It looks like we have one other guy named Jacob that's supposed to be on. He maybe doesn't know. You know what I'm going to do. I'm going to text his.

His co founder, who I think actually signed him up. This is the CEO of Boostly, which.

Is an awesome local company here in Utah. If you've heard about it.

Jacob is supposed to be on.

Okay, all right, that's done. So. All right, here we go, guys. Time to dive in. Share screen. I'm Going to share my screen. By the way, guys, if you don't know how to share your screen, learn how to do it. We're gonna be doing a lot. Okay, I'm gonna have you guys pull up your screens here just in a matter of minutes. Let's see, we're gonna go to x.com so I like you, I like you can go straight to gro.com, but I like going to x.com and then bouncing into Grok. Grok. Guys, this Grok 3 model right here is very, very powerful. It's one of the best AIs out there. It's incredibly good at coding. And one of the things I really like about it is this deep search and the think functionalities that integrate. So deep search is really awesome. We're going to do some market research later on, you know, on your business idea, things like that, and it will be able to search all of x.com everything that anyone is saying about whatever, along with hundreds of links and be able to compile a month's worth of.

Research for you in, you know, 60 to 120 seconds. Right?

So GROK is very, very powerful. We use it a lot today. But to begin, what I want to do is especially for all of those non coders and even for those coders, right, we're going to do some just coding with AI without using an integrated development environment, an IDE or whatever, we're going to just ask AI to teach us how to code something and, and then we're going to follow its instructions a little bit. Okay. And we're going to use a language, we're going to use a front end language called P5JS. And if you don't know what that means, that's okay. You're about to learn what it means. In coding, there's front end languages and there's back end languages.

Okay.

And you know, I'm going to be very, very high level on this, but front end basically is everything you can see, right? So right here we see buttons, we see places we can type, we see, you know, a black background, we see fonts and we see icons. That's all front end type stuff. Back end stuff we won't worry about right now. But we're going to use a front end language. So it's really all, you know, it's all rendering, it's all, all the logic happens all on the front end. And it's called P5JS.

It's just a JavaScript thing that's kind of fun.

So here's what we'll do, we're going to say please write the code. By the way, guys, I Write please.

To AIs because when they take over.

Someday, I just want them to remember.

That I was kind to them.

So you really obviously don't have to. I don't know if it actually makes.

A difference in the code.

I know that if you swear at your AI, it actually gets better results sometimes because it senses your emotion and frustration, it's going to think harder about it.

That's actually a fact.

So I don't know what please does, but anyway, so please write the code for a mind blowing visualization in P5Js.

And show me where to deploy the code. All right, let's see what it says.

All right, so it's going to make something cool here.

Probably look something like a screen saver or whatever. Let's try this here.

All right, so you don't have to know what any of this code means. You don't have to at all. Okay, where to deploy it? Editor.p5js.org okay, so we're going to open this up, we're going to copy the code here. And I'm going to come in here, I'm going to paste it, I'm going to click play. Okay? Boom. I just wrote some code. Super cool. All right, it's looks like a bunch of little sperms. If I'm going to be just throw that out there. All right, so let's make it, make it awesomer, guys. I'm very purposefully here giving it what I would call less than good prompts. Because I just want you to see how easy it is. Anybody can do this. Jessica was sitting at a kitchen table with us just a few weeks ago and she was telling me about this game that she wanted to build. And. And I was like, that's cool. You know, I'm just kind of listening. I pull out my phone, I just keep talking and whatever. And about two minutes later, I just said, I built your game, right? I turned the thing around and this is all I did.

Jessica.

It's very, very easy, right? I used AI to build some code. I told it what I wanted and it was done right. All right, well, it's awesomer. They added it. Added some weird thing here. Okay, so you can see that when you, yeah, it is kind of cool, it's kind of whatever. But when you don't give it a lot of direction, you know you're going to get something random. You're not going to get exactly what you Want. Okay, so now let's be a little bit more specific. Okay, Please. Okay, now please make a new.

A new thing in P5Js. All right, I'm just going to say thing. Let's make a model of the solar system that spins and is. Yeah, that spins and is true to the solar system's actual patterns. Make it interactive by allowing me.

Whoops.

Allowing me to click on the screen and have a shooting star come out of the cursor.

Yeah, let's try that. Okay, so now we've got a model of the solar system coming. So, guys, one thing I'll introduce you today is something called markdown. Like this thing right here, these little green things. This basically tells you what the code does. Whenever you have that forward, forward, slash. It's basically just kind of labeling it so that if you did care to try to understand or to learn this or just to see, like, what's going on in a certain part of your project, you can. You can know. Okay, hey, you know, this piece isn't working. Let me go and look at that and see if it's doing something. Or you can say, hey, right there, where it says. You know, where it shows lightning or lighting.

Not lightning, lighting.

Let's. Let's make that a little whatever.

You know, let's change it so you.

Don'T even know what it says again, even. But you can kind of start to understand it if you look at it.

And pay attention to it. Just try to use your brain.

Okay, so this is kind of crazy. I don't love it because the planets.

Are not circling on the lines around the sun.

So I'm going to tell it that.

The planets are spinning on the wrong axis. They're not on the lines around the sun. Fix this.

Okay, so hold on. Let's see what happens if I click. Yeah, if I click, I do get a shooting star. So that's good, right?

That's pretty nice.

Also, the planets are going super fast.

I probably should have told it to slow it down a little bit, but.

Let'S see what we get here. So it's writing some code. And.

Wait for it.

You have to wait until it's done. It actually is just thinking through it and writing it out in real time. All right, copy.

Let's see if this fixed it. Hey, play.

Okay, now at least they're on the same thing. Oh, hey, look, I can grab it and drag it.

That's actually kind of cool.

All right, now.

Okay, the planets are spinning super fast. Please slow it down so the visualization. Visualization is. Or, I don't know, palatable. Palatable. Also, please add some twinkling stars in the background and make the entire background rotate slowly.

Okay, so, guys, this is all it takes to be able to code something today, right? At least with this basic P5G is what we're doing. And throughout the entire. Throughout the entire class, this is how AI. This is how AI works. You talk to it with normal language and then it figures out the hard.

Stuff and builds for you.

And then we're going to start getting into agents that. Where you can tell it something and then it does. It tells a bunch of different things on its own and allows you to dive in and build very, very complex and amazing things. So we're going to paste this. Okay? So now we've got. We've got this.

I don't know, some stars and some things right here.

These are still going a little bit. I don't. I also don't love the contrast and the plants just so going, you know, a little fast.

So.

But I. You can see that I built something kind of cool, right? It's. It's going. It's whatever. Now what I want you guys to do is I want you all to build your own thing. So I want you all to open Grok, okay? And I want you to. And I'll go back to my very, very top prompt here, right? All you have to say is please write the code for something, right? So you could say, Evan, you could build a little golf game. Make a little golf game in P5Js, right? Jessica, you could make your flappy bird game right now in P5Js. Hey, please write the code for a flappy bird game in P5Js, right? And then say. And show me where to deploy the code. So here's the. Here's your prompt. It's on my screen here. You know what? I'll copy it and paste it into the. Into the zoom chat.

And we'll drop this into everyone paste. All right, let me just double check participants here, see if anyone's trying to join. Nope. Okay. All right, Jacob will have to join in a little bit.

By the way, guys, I am recording this, so if you ever feel like you get behind or whatever, you're going to be able to go back and watch it, pause it, do it. But for now, let's all try to.

Keep up on this Grock stuff.

So just go to grok.

Get.

Getdex.com and. And work on something, okay? And while you guys do that, I'm gonna keep working on my, my thing. If you have a question, just speak up and say, john, I don't know what to do. I don't know where to go. Tell me right, that I. Don't be afraid to say that. It's better for you to say that and for me to walk you through.

One, you know, individually than. Than for you to get behind.

Okay.

What'S up?

Our business idea currently or just something really simple?

No, this is just for fun. This has nothing to do with your business idea. We'll do that in a minute. Right now I just want you to write something cool. Copy mine. Make a solar system model. Make a model of Earth's population from the beginning of time until now and make it animated. Show a weather pattern, whatever the heck you want. Make a version or a game of a dinosaur chasing you. And you ask it to have your arrow. Use your arrow keys to run away.

Know from the dinosaur that's chasing you.

Make anything you want. It can do almost anything you can imagine other than, you know, this is not what we're going to use to.

Build something, you know, a B2B. You know, something that's soft, you know, it's like SaaS, enterprise software. That's not what we're doing here.

Make a game. Make something fun. A fun visualization. Does anybody have anything up and running yet?

I have a very rudimentary bad game of Pong exhaust.

Does it work?

My curses are too, too slow to catch the ball.

Got it. We'll tell it. Say slow down. Yeah, once you get it working, I want you guys to share your screen. I want to see what you have, okay? So speak up as soon as you have something running. Just say, hey John, I'm ready to share. And if you don't say, hey John.

I'm ready to share, I'm going to ask you anyway. I'll just pick you out of the crowd.

So I got something for you, John.

Yeah, let's see it.

Let's see here.

This is for my dance party, for my kids that they always want to do.

Awesome.

Let's see here.

Did it share?

Is that the wrong thing? Is it showing their sharing?

It started. I can't quite. It just says it started screen share.

But I can't see your screen yet.

Bizarre. Let me try.

Yeah, maybe.

Maybe stop it. Start it again. Yeah, I don't know.

I don't know why.

I don't think that's my computer's fault, but you never know.

Yeah, I think it's my computer's fault. Can you still not See it? Oh, there it is.

There we go. There we go.

There we go. We're up.

All right.

Okay.

Okay. So you built a.

What do you call it?

It's like a disco ball thing.

A disco ball. A disco ball. It was like a reflective.

It's. That's not the word. It's a disco ball.

No, I can see all the little.

All the little colors. Yeah, that's pretty neat.

Now it'd be really cool to make this a VR environment. Throw on your VR headset.

Yeah. And then you can just, like, look around. Tell it to add. Tell it to add a laser. A laser show on the ground as well.

Will do.

All right, who else can show if.

You want to see me do that?

No, no, go ahead. No, no, you can.

Yeah.

Once you have it done, pull it up.

Okay. Trevor, I know you have something up and running. That's awesome.

Okay, let's go.

Let's see.

I just barely got it working.

Oh, that's cool. Awesome.

Guys. Part of this exercise is also making sure everybody knows how to share their.

Screen because we're going to do that a few times throughout the. The clubs.

All right, so this, if you've ever played Agar IO. Yeah, I'm collecting dots and getting.

Collecting dots. Try not to. Okay. Yeah.

Cool.

You see, you have to be bigger than the dot around you, right?

Yeah.

So the red dots are computer controlled.

Right.

And they'll try to eat you if you're smaller.

Now I'm.

You've won this one.

Nice. Super cool.

I love it.

Oh, wow. Boom.

You went huge all of a sudden. All right, very cool. All right, I'll stop your sharing there.

Very nice. Evan, how's it coming?

Let's see here. Mine's kind of a lame o golf game, but here we go.

That's.

All right.

This is a lame o golf game. This is. Guys, right now we're coding for, like, the 80s. All right? So we'll. We'll catch up to the 2000 and 20s eventually.

So here's our first shot. And these. Tiger Woods. I guess that's Tiger woods, but there's something. But I'm trying to get it into this hole. I put in some trees, some bunkers. This is supposedly pebble beach, but. Yeah, so I can't get it in the hole, as you can tell.

So you can talk to it. I know you probably have gone back and forth with it a couple times. Yeah, but yeah, talk to it and say, hey, you know what? You know, how can we make this faster, better, whatever. But that's awesome. It's working. And it looks like you went straight to grok.com and you're using the. You know, you're using the in browser. I'm sorry. Or inside of Grox, you know, preview.

Tool, which is cool. Instead of going over to the P5JS editor, which works.

Right? Perfectly fine. Very cool, Evan. Thanks for sharing. Okay, let's dive over to. Hi, Christy. Good morning, Devin.

How's it coming there? What do you got?

I'm just made a game.

Cool.

Does it work?

Yeah, yeah, I think it works.

Okay, share the screen. Let's.

Let's see what you got.

I'm always awful at screen share.

That's all right. You're good. You're good. By the way, guys, this is super fun to show your kids, like, if they're bored and like, whatever. And just like hand them your phone and say, hey, make a makeup game. Right? And if. As long as they know how to type, as long as they're probably eight years old, they know how to make this.

My kids, my.

All of my kids, all the way down to my seven year old, can.

Make games with AI.

Okay. All right, let's see what you got, Devin. Is it coming up? Yeah, I can see it now. Yeah, it says R to restart. All right. Whoa. What the heck?

This thing's crazy. This is game.

I told it to create Homer's Odyssey and in a game format.

Oh, my goodness.

While.

While Homer was tripping on mushrooms.

Yeah, exactly. So what are you doing? Are you controlling one of these? I can't tell from the. Yeah, I'm controlling the white dot. Okay.

The white.

And I'm trying to avoid these kind of moving things. And then if you get down here, there's like a force that kind of pushes you up.

Oh, yeah.

So.

Man. Yeah, that is trippy. And are you trying to avoid being hit by the bigger ones too? Yeah, when you hit these guys, you die.

Got it.

And it's my score. Counter up here in the top.

Yeah, I could see it.

You're pretty good at the game already. To be honest, it's pretty easy.

I think anybody can be good at this one. That's awesome.

Very cool. Awesome. Thanks for sharing. Jessica, what do you got?

Yeah, go ahead and unmute.

Okay. Sorry, hold on. I don't know how to screen share and I haven't figured it out yet. I'm having trouble.

Yeah, just on the zoom either on. If you join via the web, then use. Just go to that zoom, whatever. And then just. There's a.

There's A go to the tab, or if you join via the app, then.

Okay, well, I don't have anything yet anyways. I gotta figure out how to deploy it because it's making me, like. It wants me to log into different ones, and I'm trying to figure out an easier way.

Yeah, I want you to go ahead and share your screen. I want to help you debug this. Right? This is. This is actually critical. So go ahead.

Hold on, Wait a second.

Hold on, hold on.

So I go to the view.

Yeah, go. Yeah. Go to Zoom. And then there's a little, like, share.

Button kind of down in the. In the toolbox or the, like, I don't know, toolbar.

There you go. Oh, wait, that's leaving me. I'm terrible at this, John, that's okay.

That's why we do this at the very beginning.

You're gonna have to let me figure it. Oh, here we go. I figured it out. Sure. There it is. Okay, so I don't even know where I'm at right now. Sorry. Can you see? I'm not even going to be able to.

We can't see your screen yet.

We can't.

Yeah, you have to push. When you push.

Okay.

I was going to say it'll bring up a. Like a.

Like a share panel, and then you.

Have to, like, click full screen, and you have to click Share again. You have to, like, click a few buttons to get it.

Okay, well, I'm. Okay. Screen. Entire screen.

There we go.

There you go. Now you got it. Right.

Well, it's almost. Okay. It says Jessica Tenney has started.

Okay. There we go. Okay, so. All right. Okay, good. So this is perfect. I want you to go back to the very, very top.

Okay. So I went.

Yeah, scroll back. Okay.

Okay, wait, hold on. So scroll. Scroll all the way up, all the.

Way up, all the way up.

I want to. I want to show you where you did just a little bit wrong in the prompt here. Okay, so you didn't specify P5Js, right? You have to actually say that language. Just like I copied over in the prompt. Right.

Show.

So please write the code for a Flappy Bird game. So all I need you to do is this. This is super easy right there, where it says ask anything. Just click there and just say, please rewrite this in P5Js. By the way, if you misspell guys to AI, it's.

I promise you it knows what it means. So don't be too worried about misspelling things D5Js. Okay, perfect.

Okay, so let's See if that works.

So then do I go ahead and click stop?

See the little stop button in the bottom right?

Okay.

Yeah. Click stop. Yeah. Yep. Hit stop.

Yep. Just so it stops writing code.

And then go back to ask anything and say, I'm going to deploy this here. I'm going to actually do.

Yeah, yeah, just.

Just type. I am going to Deploy this on editor.p5js.org P5. Sorry. Go back.

P5.

P5. Thank you. P. P5js.

All right. Not P5js. I said that wrong.

P5js. Yeah. Okay, perfect.

This. So, so, guys, I, I. It's okay, Jessica. I don't want you to feel bad that I'm, like, pointing you out. This is really, really important. Right. You just have to be really clear what you want to do. Otherwise, it's going to give you something. We're gonna be like, wait, that's not exactly what I was looking for. Right. But if you. But as soon as you're clear now it knows. Okay, I want to do this. And it understands the context of what that actually means. Okay. So now, Jessica, go ahead and scroll down and look at your code that.

It'S writing for you.

Okay.

Go back to grok. That's okay. Go back to grok.

Okay, well, I got to make sure my zoom thing's going away. There you go. Okay.

Yeah, I know that the zoom panel.

Comes down slowly and it goes back up slow. I'm sorry.

Scroll down and let's look at the code here. Let's go all the way down. Let's make sure it's finished.

This is the bottom.

Okay, perfect. Now scroll back up just a little bit so you can see at least part of the code. And then just quick copy right there, that little copy button.

Oh, sorry.

Go way back down.

You need the last. The last prompt. This is your old one option. Okay, you scrolled up too far.

I'm going to go all the way down.

Yeah, all the way down again. And then just scroll up a tiny bit.

A little bit. Like a copy right here.

Nope, nope. You need to get. Yeah, scroll up just a little bit more. Little bit more till you see your action, until you see the actual code snippets.

See the code.

Keep going.

This.

Yeah, here's snippets. But I don't know.

I feel like if it was, like, super long. Scroll all the way to the bottom.

I don't know. I know I went all the way to the bottom. This is not it. I think I need to just ask it to do the whole thing.

No, no, no. You need to go. You need to go all the way to the bottom. Nope, you're still in.

You're still spot. Yeah, there you go.

Now go back up just a tiny bit. Yeah. I think it's because your scrolling function.

Is just going really fast.

I'm going too fast.

Yeah.

Okay, here we go. We're bringing in Jacob Thompson, who is rolling now.

Hey, welcome, Jacob.

Thank you very much, John. Hey, everybody.

Yeah.

All right.

Jacob was waiting till the morning, but I'm glad we did that. Okay, Jessica, we're almost there. We got this. Scroll up just a little bit until.

You see the code. A little tiny bit, I'm telling you.

Right?

Keep going. A little bit more.

I have to go.

Yeah. I don't know what all this debugging stuff is.

I think it's because I hit stop. Do you think it's because I hit stop and I don't know how to finish?

I don't know. But I'm going to have you. I'm going to have you stop sharing your screen for a second and.

Okay.

What I want you to do, and then I'll figure it out a little bit. Jacob, to catch you up very, very quickly. Right now we're coding in Grok, okay? So if you want to just open up grok or go to x.com and click on grok, and we're writing in a language called P5JS. We're almost done with this, so you might kind of miss out on this. You'll be able to go back and watch the recording later and do this part. It's pretty.

Pretty fun, pretty easy.

Does anybody else have something they want to show that has not shown yet? I think we've gone through pretty much everyone, Eric. We don't know if we. Did you.

Yeah, yeah, I've got a cool one.

Yeah, go ahead and share it. So, Jessica, what you need to do is you need to maybe just ask it again or maybe start over the prompt again, Right? Just start a new conversation with Groq so you don't have all this crazy stuff and just say from the beginning, make sure you get that prompt right. So, Jacob, if you just copy the thing I just dropped into the. Into the chat there for everybody on Zoom and just put that into Grok and then, you know, build something cool and then click copy on the code and drop it into the visualizer there. Okay. All right. So what do you. What are you building for us here, Eric? Tell us about it. All right.

It's pong my left mouse moves this side, then my up, down, keys. So I'm playing as myself, and if I hit a ghost, I lose a life. And then three lives declares the winner.

It's like Pac man plus Pong.

Exactly, except for there's no more ball. So I did not write that in there.

Now it's gone. What you would do to fix that is you'd say, okay, the ball disappeared.

But the game didn't end right.

Then it would know, okay, hey, if the ball goes off the screen, restart the session or have them push R for restart or whatever you want. But, guys, the purpose of this, and Jessica, I know you have maybe I want you to keep working on it for a second here, and then I want you to just say, hey, I got it working.

Take.

Okay, after you follow that prompt a little bit. But what I want, what I want. The. The purpose of this section here is really to show you guys that you can all code something. And you all just did it. Right? Like, Evan, I've. I've. I'm sure you've never written any code ever before. And to be able to, you know, jump in and build, while it's very rudimentary, a quick golf game is pretty cool. And I promise you guys, if you were to spend. If you were to spend, I don't know, two hours sitting here prompting this thing and refining it and fixing it and make it more complex and making a start screen and making a leaderboard with scores and, you know, all these different things, you could build something pretty dang cool in just a couple hours. Right? It's amazing what happens when you start.

Refining and pushing deeper and deeper into that.

Okay, all right, let's. Let's move on. Any questions about that before we move on?

Okay, so quick question.

Oh, yeah, go ahead.

Not being a developer, why would I push this out to editor P5JS? Why would I use that instead of it? Because it looked like it was just doing it in grok for me. Yeah, but what's the point? Going out into the editor?

Yeah. What I'm trying to teach there is that if you don't know how to deploy the code, AI will tell you.

Right.

And you figured that out on your own, right? You figured out a new way to do it inside of there. But you can. There's a ton of. Ton of visualizers of every type of code where you can run Python scripts and you can run JavaScript full on, whatever you can do 3js, P5js. You can run any code. You can Imagine out there. Heck, I'm sure there's probably some Flash.

Players out there still that you could code for and play around with if you wanted.

But yeah, bottom line is you can open up an AI, ask it to write some code, and then you say, what do I do with this code? Oh, we put it here.

Right.

My daughter, who is 10, built an amazing game the other day and she did it all in JavaScript and she did it in a fairly complex way where she had to copy over the HTML, the JavaScript and the CSS into a specific type of viewer. And she figured it out. She just like, oh, I just go here. Because all she did was say, how do I make this run? Oh, go here and do this. And she just followed the instructions and.

She had it up and running in 10 minutes.

So, um, anyway, it's very, very easy to, to iterate very quickly. It understands your context. Jessica got a little lost in that one because her, she gave a bad prompt in the beginning and then it gave all kinds of unnecessary stuff and then she just got a little bit confused because it was a lot of unfamiliar stuff. And, and anyway, so that happens sometimes.

Guys, what's the prompt we were supposed to use?

So the prompt, if you look, this, this will help too. Jessica, if you go to the Zoom, there's a chat in Zoom, right? So if you go to the Zoom panel, just like where you found share, you'll see a little chat. I'm going to drop a lot of prompts into the chat to everyone. Okay. So it's just a message from me to everyone. You just copy it, paste it in. You know, I copied my mind blowing visualization, whatever thing, right.

And.

But you can, you know, and then you could just modify that and say, okay, instead of mind blowing visualization, I.

Want my flappy bird game or whatever it is.

All right. Okay. Jacob, did you get something up and running?

Just curious.

I put in the prompt and what I got was like a fancy background screen that kind of dances around.

Like a Windows 98 screensaver or something.

Yeah, yeah.

Anyway, that's awesome.

I don't know if you're alive in 1998.

I don't know how old you are.

But you look young, let's just say that. Anyway. Hey guys, right now we're going to dive into some intros. I want to do this quickly because we have nine people and like I said, three hours goes fast. Actually have eight people.

Well, I'll introduce myself as we keep.

Going, but what I want is your name. Where do you live? And then, and then we're going to dive right in here, guys, to your business idea that you want to do for this course. Now, before we do this, some of you guys might have existing businesses, like Jacob, you've got Boostly. I know that. Right. Maybe you want to build a tool inside of Boostly, or maybe you want to try to rebuild a feature that you already have and see if you can rebuild, like something that you paid, you know, $400,000 to developers to create, right? And be like, guys, I built this in like a day. We can, we can go faster, right? I feel a little bit bad, but I've had many CEOs on this course. By the end of the course, by the end of the three days, they.

Have fired some of their developers from.

Their company because they go and they talk to them about it and those developers aren't willing to work with AI. Or they're like, I can't do it. And he's like, no, I did it. And you should be able to do stuff way faster. And if you can't, if you can't flip that switch, we need to move on. Okay, so anyway, I'm not saying we need to fire developers. I'm saying developers need to learn to work with AI in a big way or you're going to fall behind. All right, so if you don't have a business idea yet, that's fine. You can say, I'm not really sure, or if you just have an idea, we'll work through it. Right now you get 45 seconds per person.

That's the goal.

Okay, so quick intro. What's your business idea? And then I might have a little bit of feedback as we go, but let's do that. Trevor, go ahead.

All right. I'm Trevor Studdard. I live in western Utah. My business idea is kind of a motivational software as a service platform where if you've ever used something like Nectar or Motivosity or any of those tools. Yeah, those seem to like. There seem to be like a kind of like a Twitter with points. I'd like to do something a little bit more that's more like of a management tools where people can kind of like post what we've been working on. You can skip more scrums, recognize people better, have groups of groups within groups.

That's.

That's kind of the base of it. Base things I need is like ability to like, give people gift cards, give people other ways to buy, like company merch and stuff like that.

Love it.

That's cool. I'M very familiar with motivosity. That's awesome. Take them down. I love it. Brian, what do you got?

Hi, I'm Brian from Northern California. Basically my idea is to basically make like some kind of scraping tool. I used to buy and sell a lot on Craigslist, Facebook, that kind of stuff. Just got a car wreck, so I messed my back up pretty good. So currently not employed. So move forward. Want to kind of find something to make some side income. So my idea is to make some kind of like scraper so I can go through Facebook, Craigslist, that kind of stuff and find items I'm looking for without having to manually scroll for hours looking for a certain kind of item.

So are you trying to find imbalances in the market where maybe one platform is selling it for $20 and another one's buying it for $40 so that.

You can then buy and flip type things quickly?

Yeah, the idea, that kind of arbitrage.

But then also just the fact like, you know, so especially bigger items are hard to move because of shipping. So if I can find it and you know, if it's a three hour drive, whatever, I'll go take that drive and then you turn around, sell it back on eBay for five, six times my, my cost.

Nice.

Very cool. Love it.

The, the searching part because that takes a long time and then in the future maybe I could develop, you know, posting and selling that, that side too. But basically I wanted to find the best item so I don't have to worry about it and then alert me when it finds an item for me to manually review and see if I.

Want to pull the trigger.

Amazing.

I love it.

Travis.

So my name is Travis. I live in northeastern Georgia. My business idea is there's this space. So it's social media marketing, basically. So I'm trying to find small influencers with less than like in the 25,000 range. And then the idea is to find people who have never been sponsored before, connect them with small businesses who want social media marketing. But then the small businesses don't have to pay a fortune to a social media marketing company. They can just do it themselves for like however many dollars a month and use this platform to, to just, you know, and then it'll draft the messages for them. All they have to do is copy and paste and DM create it and say, hey, I'd like to sponsor you, let's see if we can do a collab, that kind of thing.

I love that.

I've got a lot of ideas around that.

All right. I need to remember that.

That's very cool, Eric.

So I'm looking at creating in addition to a custom GPT that I've created that is kind of like a therapy bot. You can take a one hour session and it goes through, systematically finds out what your self advertising beliefs are in an area of life and goes through a five step process to remove that. And by the end of that hour you have the new belief, a new action that is inconsistent with your old identity and then steps to actually implement that in your life. But I would love to take that off of ChatGPT and run it through a system so I can actually get access to people's conversations so I can then improve it. And then ultimately I would love to give the tool for free and then have it do advertising within the experience to different platforms that I'm on or different cost coaching, masterminds, retreats, type things that I offer. So essentially it's a free tool, but then you're getting advertising for. Hey, would you like to do a retreat for business owners that are also trying to overcome these types of things?

Yeah. And so, yeah, I love that. So, so.

Okay, well, I love what, what I, what I heard from you is the first AI powered app that we're going.

To be building, right? And that's really cool.

We're going to make something AI powered for you and you're going to, you're going to make it.

I'm going to just guide a little bit at a high level. But, but that's awesome.

And guys, if you have AI that.

You need to integrate, it's, it takes like, I don't know, three minutes to integrate AI into your app. It's super, super easy via what's called an API. We'll, we'll teach about those later. Jessica, go for it.

Well, right now I'm trying to figure out how to build a flappy bird game, but after that I would love to build an app that is also kind of along the like positive health. I'm a nurse by trade and I would love to build an app around how to help people, positive affirmations, and how to incorporate positive thinking into your overall.

I love it.

Very cool. And where you, and where do you live?

Sorry, you skipped that one.

I live in South Jordan, Utah.

Awesome. So we've got, we've got a West Jordan and a South Jordan.

We'll see if we find a North.

And east somewhere in here.

All right, Evan?

Yeah. I'm in Highland, Utah and my idea is a little bit more simple. I started A tourist business over in Hawaii. I think there's just a lot to disrupt in the travel industry, just in general in Hawaii. Spend winters there and done a lot of research, learning about just how things work. And so honestly just looking to use this to, to build a website, look at the business from different angles, how I can, you know, create some type of customer service platform and just automate a lot of parts of the business where I don't have to hire hourly employees, which is not the best thing to do in Hawaii. Expensive and not terribly reliable. A lot of times probably at the beach or surfing. But anyways, yeah, so mine's simple, just building a website, marketing it and figuring out how to get the customer service side and some of the other business operations a little bit more streamlined with AI.

So I love it.

Automation is big time. AI can obviously think like a human.

And present itself like a human and it can do incredible things. So making chatbots and things that can answer and you know, think QA on crack, right. I mean, you know, questions and answers. Right. You can, you can have, you can have that thing just super interactive. And then it can do things for your customers and book out your systems without you ever having to show up for work so you can go surfing.

I get it.

That's awesome.

Devin.

I wouldn't even know where to start. I've got like five or six different use cases, so I'm not really sure which, which direction to take it. But I have a SaaS company that serves people with intellectual developmental delay. So there's one feature that I'm really interested in building out particularly it's like a voice, a voice to chat AI feature. So people with disabilities, they, they consume a lot of labor time for these care facilities, these caregivers within facilities, and they're surrounding like really basic conversations about like favorite movies and whatever. So I want to create like a therapeutic robot, if you will, for these people with disabilities. Like have somebody to talk to, to kind of answer their questions because AI won't escalate in their lack of patience over time, which ends up being counterproductive to the disabled person. So kind of build out. I have, have a lot of it through my developer group kind of built out, but I want to be able to like build a model, flush it out myself because I'm not able to get my dev team to truly understand its use case. And so I either need to build them like a really good spec or.

I need to be able to like.

Build it most the way myself. And like, no, it has to be this way.

Yeah. So yeah, I absolutely love that use case. Devin, for those of us I know.

There'S a bunch of entrepreneurs on here.

And, and I'm right there with you.

In fact, I think everybody that's sitting on here is an entrepreneur or trying to improve their skills as an entrepreneur. But one thing that has held us entrepreneurs back, right, is we gotta hire expensive people like Trevor, right before we can see anything, right? At least up to now we've had to do that and I've had to pay a hundred thousand, two hundred thousand, three hundred thousand for crazy senior devs to do, you know, crazy stuff out there for me. And, and now we can have something, and I can have something up and running in 30 minutes, like something really good. I'm, I like everything I'm going to teach you over the next nine. Nine hours. Over the next three days, right? Three. Three. I can do it about 30 minutes and, and I can have now, I, it takes a while to refine it now, right? When I, when I did my, you know, exercise of building a business, launching a business in three days, I spent about 30 hours on it, right? Because I really needed it to be good. It needed to be a complete experience and I'll show everything I did there. But, but the reality is I built, I wrote, you know, 150,000 lines of code that's pretty dang good code because I have tools to check on that kind of stuff too. And I know what's, I know what's going on a little bit around the code and, and I was able to, to build something that, that actually provides.

A lot of value to real businesses.

So anyway, okay, Jacob, last, last but not least. Awesome.

Yeah, so I'm at Boost Lee. My name is Jacob, I'm, I'm from Provo, but I work in Lehigh, Utah. So basically I'm, I'm, I just got promoted to a position called Enablement Engineer. So just enable individuals, teams and company be more effective. Trying to reduce headcount in the future, but also increasing expected output. So KPIs, revenue per employee, et cetera. So excited to learn how I can automate some systems that we have here and build and design them.

Okay, very cool.

Jacob, how many or how much money.

Is Boostly raised now? At this point?

22 million.

Yeah, guys, Bruce Lee is an awesome.

Up and coming startup in Utah doing.

Some really cool things, by the way. I only wanted to ask that because I know Boostly. I knew, I knew what the answer.

Was, but I wanted Jacob to brag for A second.

So I'm really excited about this. I know Shane, the CEO there and.

He'S, he's really awesome.

I said this at the very beginning, Jacob as well. But I'll say it again and reiterate here, guys. I raised $13 million for my last startup and it sucks raising money.

It's not fun, right?

Anyone who's raised money, number one, it takes a ton of time. Number two, it brings a huge amount of stress and pressure into your life as the entrepreneur. And number three, you just don't need to. When you're at the beginning of a startup, right, you can go borrow money from your dad if you want to, but then you owe your dad or your uncle or whoever it is that you borrow money from. And now you've got this weird relationship there and you know, hey, does it, did they lose money? Do they make money? And you always have this kind of thing there. You guys can build everything that you're going to build this weekend or this, this, sorry, not this weekend, this beginning.

Of the week here.

In the next three days, you're going to do it for less than $100. Like for real. And that's crazy. That's just crazy that you can start a business and get it to the point where you're going to have it. And, and by the way, the only reason it's going to cost you $100 is because you had to do three monthly subscriptions with, with, with Grok and Chat, GPT and Replit that add up to about 85 bucks or something like that. And you know, so you could build five businesses or 10 prototypes in the next two weeks. And it's still, you're only at 100 bucks. Okay? So we're in a different world. The paradigm has shifted and it's, it's, it's crazy, right? So a little bit about me, guys. I'm John Cheney. A lot of you know me, I live down in West Mountain, Utah, which for those of you don't know, I'm staring at like, I can see the Y and Timpanogas.

I'm at the south end of Utah Lake, basically, right.

So we have a little five acre ranch here and we, you know, I mostly spend my time weeding and trying to keep up with the chickens and have wonderful wife and four kids and you know, we've got a wonderful life. As I've mentioned, I founded a company back in 2016 called Seek. We later rebranded to Okabu. We had a couple exits, we had two kind of, kind of a Two part exit where we sold off kind of half the business in 2021 or 20, no, sorry, 2022 and then we sold the rest off last year in July. I've been gone from the company since that first exit.

I went with that acquisition and, and.

Have been working on some things since then, which by the way, I'm completely, literally, guys, all of the code. We raised $3 million for that new business and I'm completely scrapping every line.

Of code that has been written for that business and we're redoing the entire thing because of AI. There's some discussion we'll have around that.

But anyway, I've been an entrepreneur my whole life. I've been in sales. I've, I've raised a lot of money. Again, I raised 13 million for the one company, but I've raised probably a total of about 50 million for different companies. So I have a lot of experience with that. And what's crazy is right now I really don't think I need to raise money for a long time. I just don't. I feel like the world has changed in a way where you can build something and get it to a cash flowing position within days, right? If not, if, you know, maybe you want to work a little harder and get it going within weeks. But, but the bottom line is you can build something of value almost instantly now, now what does that mean for the rest of the entrepreneurship world that also can do this? Right? That's a discussion we're going to have. But, but it's, it's an interesting world.

That we live in.

So today the primary, the primary thing that I want to, I wanted us to work on is we're going to start, we're going to jump into chat, gbt, I'm going to share my screen. We're going to start ideas and refining our business ideas and, and using AI.

To do that, right?

Using AI to get the name of.

Our business, the feel of our business.

The ideas, the ideals, our unique selling proposition, right? All those different things. We're going to get it all done and we're going to do it in a matter of, of minutes, right? I mean you're going to have like 12 minutes to do everything I just.

Said once I say go for you guys, right?

The idea here is to be able to work fast, to use smart prompts and then we're going to do some really cool stuff. Now we're going to jump in. So I'm going to jump in now. I'm going to share my screen. We're going to jump into Chat GPT. Okay. You can use Chat GPT on a browser like this. You can also use it down here on a.

Sorry, dang.

Zoom controls are always getting in my way here. You can also use it on an app like this. There's a lot of really neat things that we're going to look at inside of Chat GPT, but one of the things that I really love, and I'm going to use the browser today because I know a lot of you will be doing that. So I just want to make sure. I'm trying to show you exactly what we're doing. But one of the things that I really love about ChatGPT is how good it is at dealing with context that.

It remembers about you.

So the context in a context in an AI has two meanings. One is how much context can you give it? Well, meaning how many tokens can you throw at it at once? Or how much text can you throw it at once? So if I were to give chatgpt right now and feed it, you know, the Library of Congress and every word that's in there, well, it's going to get overloaded. It's going to be too much information. It can't process that much. Not yet. In the future, I think we will be able to do that. You combine quantum with AI and it can do that. We're not there yet, though. Right now it's going to have a limit. Now the limit's still pretty dang big. You give it a 90 page legal document and it can, it can deal with it, right? It can, it can help you work through it and find problems and find.

Inconsistencies and, you know, it can be.

Your lawyer for you, which I, which.

I use it for quite a bit.

But it has a few different types of context, right? So one of the types of context that it has is it has a memory it actually knows. It remembers as long as you're logged in. Right. It knows who you are a little bit. And I'm going to show you what that looks like. Okay? So right here I can click on the, on this right here I can click on Settings, by the way, for, for you companies or people that might be worried about their privacy, I want to show you this. You've got a thing right here that says data controls. And then you have this thing right here that says improve the model for everyone. On if you are working for something for your company and you're worried about it getting out to the world for whatever reason. And by the way, the way that it's set up, it's really not, it's not really a concern. But if you're like, I want everything to be private, perfect. Click here and switch it off. Okay, done. If you want to include, you know, I, I personally, because I know how they're hashing their data and parsing their data and things like that, I'm actually not worried about that for me, but if you are worried about privacy, turn that off. And now you're in a completely encrypted environment, just like your, your Apple phone, right? You could text people things back and forth. You don't have to worry about Apple reading it. They can't read it. It's encrypted.

That's the same thing here. Okay?

But back to the memory personalization right here. You've got a little button that's kind of hiding here, but it says manage Memories. And if I click on this, it knows things about me. It knows I'm an entrepreneur and a composer and a consultant. It knows I'm fluent in Spanish and Chinese. It knows I love to kayak. It knows I'm a member of the Church of Jesus Christ of Latter Day Saints. It knows that I founded Seek and which later rebranded to Akabu. It knows a lot of stuff about me right now. The way it knows that stuff is because I told it, right? Or I've talked about it a lot and I could tell it, I could say, hey, I really like chimichangas.

Please update your memory, okay?

So I can tell it something and tell it to remember it. Please update your memory. Right? Updated saved memory. Now, if I bounce back into my settings here and I look at my data control, sorry, personalization. Manage memories. Likes Jimmy Changas and loves tacos, right? I gave that same example last week. I'm going to click here and I'm going to actually have it forget it, because I don't really need it to understand that context for me. Okay? But anyway, see how easy that was? I told it something, I told it to update it, and now it will hold on to that memory until I delete that memory. Okay? Now I can go and I can manage it. I could clear the memory if I wanted to, but I actually find it very, very useful for it to understand who they're talking to, right? Because when I ask it something, it will say, oh, John does this, and he likes this. And this is usually how he likes me to answer my questions, right? So that's one piece of context. A second piece of context are these conversations, right? So if I were to come in here and, and look at, well, I don't know, something else I've looked at before. Lord of the Rings animation request. Okay. I asked it to do some, some, some things, right?

I had it turn this image into.

A Lord of the Rings style animation. Pretty cool.

So made this image and then, and then did this.

Okay. Now I could come back to this conversation and I could say make another image like this. Okay. So like this, I didn't give it any instructions, but it knows and remembers and analyzes the rest of the conversation.

Okay.

And so you just pick up right where you left off. So now we've got two levels of context. We've got my, my, you know, the memories, then we've got a conversation context, and then we have a third type of context, a third layer and you can keep adding layers. There's actually a lot. Context is so critical to understand as you're building things and as you're working with AI to understand. What does it know, what does it not know now? What it does not know. And this is a lot of people, a lot of people get this confused. A lot of people actually think that ChatGPT remembers every conversation you've ever had. That is not true. It only remembers what's inside your open conversation. And then sometimes it will take those past conversations and it will maybe put a one sentence, something inside of the.

Memories here that we were looking at.

But that is all it remembers. We'll let this animation request go. We don't have to sit here and babysit it. What I love and what all of you guys are going to do now, as soon as we get going here, and you can even do it right now, if you're good at following along, if you're good at split screen stuff, do that right? If you're not, then just watch and then dive in. But this project feature is awesome. Grok doesn't have this. Anthropic doesn't have this. Some of the other tools out there don't have this. But I love the projects feature because I have so many different things I'm working on, right? You know, rec Gard weather app, I actually build, I build a new app, a brand new app. With every single class that I teach, I come up with a brand new idea on the spot. We come up with it and then, and then I, and then I move forward. Okay. And by the end, I have a working app that I could if I wanted to pursue that business. Okay, look, it gave me another image kind of like this gave me the next step, right? He's moving forward. He's ready to go. Although he did shift his helmet from.

One hand to the other. So that's interesting.

Okay, so projects, new project. Okay, so now here, I need to come up. This is going to be it, by the way, what you call it, it doesn't matter. I need to come up with a new business idea for this, for this class. And I'm thinking Evan kind of inspired me a little bit. He's got his tour business in. In Hawaii. I want to do something I've kind of had in my idea. Like, it'd be in my head. It'd be really fun. So I'm a whitewater kayaker. I love to kayak, love to get out and go on adventures and do all kinds of fun stuff like that. But kayaking's a big one. But I want to build a business where I take business leaders on adventure retreats. That's. That's as far as my business idea planning has gone. Okay? So hold on. Let me turn on my do not.

Disturb so that my brain doesn't get distracted every time my phone or my watch, you know, vibrates at me.

All right, so we're going to call this Adventure retreats. Great project. Okay, so now, guys, we've built a new project. We've got new context. Now. We have. Now we're going to have several levels of context in here. We're going to have our memory in chat GPT about us. And if it doesn't know you yet, that's okay. Doesn't matter. It's not going to affect you right now. But over time, ChatGPT will get better. The more it gets to know you, the more you talk to it. Okay? And then we're going to have context inside of this project. Now, anything we drop into this project, it's going to remember or reference or it's going to, you know, it's added into its knowledge or into its context. You can add files, right? So maybe, Evan, you have a, you know, a list already of things that you've already put together. You have some documents, or maybe you have a founding document, or you have your LLC or whatever it is. You can drop all of that stuff into your files. And now it's just kind of here, and it just knows, okay? Now it knows the exact llc. So if you say, hey, can you. Can you create a new, you know, liability waiver for me? You just say that, and it knows, oh, it's for this project. It's for this llc, because I can See that in my project here, and it's for this and this and this. It asks all those questions and answers all those questions and builds it immediately and it's ready to rock. You can also add instructions for how you want this, this GPT or this project to respond. And actually as this GPT. So this is important. We're going to go over this right now. So for this one, we're going to say for every, let's see.

Always address me as your highness.

Okay?

And.

I want you to act like.

Steve Jobs and help me build an incredible vision and mission for this project.

I want you to challenge my ideas.

And don't, whoops, don't. Just be a yes man.

If I can improve an idea, tell me. If I have a bad idea, tell me.

Really help me build something powerful.

Okay? So I've given it a few instructions. So it's going to consider, it's going to consider these instructions every single time I ask it anything. Okay? So now I'm set up. I'm ready to build a business. Okay? So I'm going to dive right in and I'm going to say, I want to build a business. Let's see. I love to. Actually, I don't even need to say that, but I'm going to say it anyway. I'd love to kayak. And you know what, I'm actually doing.

This a little bit different. This is going to be fun.

I'm going to use the voice dictation here because it's actually a little bit faster and very powerful.

So here we go.

Well, I want to build a business that really revolves around my love for adventure and whitewater kayaking and entrepreneurship and coaching. And so my idea is to build basically adventure retreats where I take groups of people on awesome. Some adventures. Maybe we're going to go, you know, whitewater rafting down the Grand Canyon or we're going to go kayaking in Mexico or we're going to, you know, head out to North Carolina and do some kayaking at the Whitewater center there. But bottom line is I want to put this together and I want to, I want to build a business plan.

So let's go ahead and get started.

Okay, so that's what I could have typed. Of course, now I just said it. By the way, I didn't have to be super careful with the way I was talking chat. GPT and AI is extremely good at pulling out the important details, reorganizing it and then, you know, putting that out. So I'm going to go ahead and give it my, give it my prompt here. And I basically just said, hey, I'm going to build a business.

This is kind of a high level idea, right? This is all we've got.

So it's going to come back and it's probably going to give me a little bit of a business plan. All right, your highness, this is a powerful idea and it has your DNA all over it. It blends your superpowers, world class adventure skills, a passion for personal growth and entrepreneurial leadership. Also, ChatGPT is going to make you.

Feel really good about yourself, you know.

So it's going to say all kinds of nice things.

Now, I obviously gave it a stupid.

Instruction to have it call me your highness, but you can have it call you Evan or call you Jessica or call you Devin. Okay? So anyway, vision, mission, target audience, experience, design. So it's starting to kind of build the business already. And I asked it to, right? I said, hey, I want to build a business plan, right? So let's go ahead and get started. So we'll let it answer, we'll let it, we'll let it finish what it's saying. All right? Premium ticket price, 5 to 15k per retreat. Tiered pricing, general admission versus VIP, one on one, better accommodations, ongoing coaching, upsell. So it's starting to give me some ideas. Now if I'm actually building my business idea, I want, I want to read what it says at least I want to, I want to consider what it says because it's going to start sparking ideas. Okay, but the thing that we're going to do here is we're going to start like this. We're going to say, all right, and this is, this is maybe your first really, really cool thing you're going to learn today about how to use AI, especially ChatGPT.

It does it really well.

Although GROK does it well, a lot.

Of places do it well.

I want you to interview me one.

Question at a time about this business. I want you to ask me about the brand or.

Sorry, yeah, the name, the branding, the feelings customers should have.

My USPs.

Marketing.

Plans, monetization plans, and anything else we need related to this project. Go ahead and start.

Okay, now I'm going to copy this. It's not perfectly written, but I'm going to drop it into the chat there so you guys can use this at the right time. But now I'm going to go ahead and click start. Okay, so now it knows. Okay, we're going to flip this script here. We're going to now start interviewing the human. Now we have the AI interviewing us instead of us asking the AI to do everything. Okay, so first, what do you want to name this business? I don't know. Do you have some ideas?

Yeah, let's see. Okay. Names that Carry Soul, Movement and Mystique.

Adventure and Flow.

Eddie Stone. The Flow Line. Confluence, Undercurrent.

White Cap.

My Cap.

White Cap's not bad.

White Cap, White Cap, Whitecap Adventures. It's kind of fun. I'll probably just do Flow State Adventures because that's, that's. I have Flow State Leadership. Flow State Adventures. That could be fun. I'm just gonna do Flow State.

I probably do that. But it gave me some ideas. I actually kind of like Whitecap. Lots of good, you know, visualization there. I think that's really nice.

You know, pass it.

Passage. Passage could be good because you've got. Yeah. Both the trip and a rite of passage. It's actually pretty decent. Right. Passage retreats. Like you could, you could easily do that. So the point is, even if I had an idea or didn't have an idea, I can just say, I don't know. What do you think? Oh, yeah, I like this. Hey, let's roll with that one. Right?

I'm actually going to roll with Whoops. With Flow State Retreats since I already have a company called Flow State Leadership.

All right, so now it knows.

Understood.

Okay, next question. Brand identity and emotional resonance. Right? How do you want people to feel? Right, so again, guys, I'm going to go through and I'm going to actually answer these questions and I'm going to keep sharing my screen, but I want you all to now dive in. Okay? Open up Chat. GPT. You're going to start a project. Call it whatever your idea is, right? Just high level. Just give it some descriptive name and.

Get started.

Any questions? Okay, I'm not hearing anything, guys. That's good. Jump in and. And get started. So your idea is you're going to first tell it just high level. And it could be two sentences, right? Evan, you could say I'm, you know, I have a business out here. It's called this. And I'm going to work on a better business plan and build a website for it. Awesome. Okay, perfect. Okay, Jessica, do your thing about your emotional, whatever, resilience thing that you're doing. Same with Eric, Right, Jacob? Hey, I'm working on these tools. This is what I want to do, right? Just tell it exactly what you want to do and then say, hey, I want you to work on a plan. With me, our business plan in order to build this business or build, build what's going on here, right? Just start talking to it. Then you'll use that prompt that I, that I dropped there in the, in the chat to get it to interview.

You about this idea.

Okay?

All right.

Feel free to jump and ask me questions if you can't figure this out. Guys, it's fine.

Tell me.

And don't forget, guys, also, when you set up the project to give the GPT instructions to say, hey, I really.

Want you to challenge my ideas, right? I, I like, I like adding that.

Otherwise it will be a yes man. It'll just say every idea you have is awesome. Or I mean it, it will, sometimes.

It will, it will question you.

But I, I like to explicitly tell.

It to challenge me.

I'm just going to keep working on my questions here on the screen and.

Ask, you know, jump in and speak up if you need something.

And guys, we're going to take till 10:25 to do this. Okay?

You got 11 minutes to build your business, so just answer quick, do your best.

You can adjust some of this later, so don't worry a huge amount. But I want to show you the process. Okay? So guys, I just, as you're doing your thing, keep going. But I just asked it a question. So it asked me, hey, you know.

Why are you different? Why are you different than other retreats, right? How are you going to really stand out? And what I said back to it.

Is I told it a little bit, but then I said, hey, where do you think there is opportunity in the.

Market to improve upon what's already out there, right?

And so then it will take a minute and we'll just kind of go.

Out and research a little bit and then come back and say, hey, you know, here's where you can be different. Most people do this, you can do this. Most people do this, you can do this.

And you can start to really create.

Something that's differentiated in the marketplace with just, you know, one question, one prompt, really powerful one there. Where do you think there's opportunity to improve upon what's already out there? Ask it.

And then say, yeah, I know, I.

Love this, I love this.

And you don't have to agree with everything it says. Maybe you say, you know what?

I want to focus on point 1 and 2. And, and let's, let's keep moving forward with that.

Has about five more minutes.

We'll be done with this first step.

So as you're going through the interview questions, you should get through Maybe, maybe.

Seven or eight of them.

Once you've gotten through about eight of.

Them, just kind of chill and, you know, you might. You might still need a few extra minutes to get even to that, but do your best.

All right, how's everybody coming? Who's done?

Raise your hand if you're done.

Raise your hand if you would like another five minutes.

Probably need a couple more hours.

You got five minutes. Okay, we're gonna go till 10:30. All right?

Yeah, go ahead.

Go ahead and do that.

And yeah, just keep.

Keep talking to it. This is good, guys, right? You're thinking about your business in ways you probably haven't before.

This is awesome.

Hey, John, do you use 4 or 4.5 on chat GPT? My 4.5 is going quite slow.

Yeah, I'm using 4.0.

Yeah. 4.5 is probably going to think too hard for what you're doing right here.

You. You really want more speed on this. It's not bad, right? But 4.0 is more than enough to do this part.

It's not having to think through really.

Complex things right now. And that's really what four or five would be for. It's interviewing you coming with ideas. Right? You're doing the hard thinking right now.

By the way, guys. If the interview ever, like, gets off.

Track and it just starts going somewhere else, say, get back to the interview. You just say that, and it'll remember.

What you were talking about earlier that.

You asked it to interview you.

All right, guys.

All right, Devin, it looks like you just sent me something here.

Got it. Is it.

Is this what it just asked you, Devin?

Yeah, it just says, hey, what do you. What do you want to do next? And so I'm.

I kind of took a little bit.

Different route just because this isn't. And I could shift gears back to, like, more of, like, a startup business from zero. But my idea was to kind of build an application that's embedded within our existing company is existing company software. So I didn't know how to, like.

Blend the new business.

I mean, I could turn this into.

A business on its own and just.

Build an app that does this.

But it'll lack data, and I already have existing, you know, data to pull.

From to train the AI.

Sweet.

Awesome. Very, very cool.

All right, guys, so we're there now, and if you're not done, go ahead and keep, you know, talking to it a little bit. But after you've gotten through about seven or eight questions, it. It kind of understands what's going on here, and you've Read some stuff. And you said, yeah, do this, let's do that. And you can always go back and refine this later. Now, when I built my business in a weekend, I took two hours doing this step. I went deep into it, okay? Because I really read every word and I thought about it and I had to do research and I bounced back and forth. And we're now going to do that. But the next step here, guys, is to say, I want to build or build me. Just say, now build me a 1500.

Word business plan based on everything we've discussed.

Okay, Go ahead and type that into your. Into your team. Right? Build me a 1500 word business plan based on everything we've discussed. And it should open up this Canvas thing like it just did.

If it didn't, and it just does.

It right in the chat, you can say, hey, please make this in Canvas. Canvas opens up like a document creator.

On the right side of the screen, and it's really useful and we'll need it.

So you want it to work in Canvas. Okay.

So go ahead and, you know.

Now build me a 1500 word business plan in Canvas based on everything we've discussed.

Okay.

Chat. This is an.

My web page for my existing software.

I know the web page needs work, but please meet. Yeah, I muted you, Devin, Just FYI.

We were. We were going to enjoy your. Your thing, and I love that you're using voice.

That's awesome.

John, just so I understand correctly, you want us to get that 1500 business, the summary and have them do it through Canva Canvas.

Yeah, not Canva. Not the. Yeah, not the Canvas. A totally different platform, but right here, the Canvas tool.

Yeah.

Now build me a 1500 word business plan in Canvas based on everything we've discussed. And it will open up just like my screen splits here. If you can see my screen, it.

Opens up a document kind of thing. Right, Cool.

Thanks. Yeah. So, guys, that should only take a moment, right? Once it's going, it's. It's going. Okay, now what we're going to do is something cool. So how many people have their business plan done and up and up and rolling? How many people do not. I guess.

Let me ask that. Okay, Jessica. Go, people. All right.

Most of you should have this done. Where are you at? What. At what step are you?

Jessica, how close are you? You're muted. Sorry. Y. You guys will be expert experts at zoom by the end of the course.

Yeah, I'm close.

Okay.

All right.

Okay, good. So again, build that business plan. Here's what we're going to do next, guys. Now we're going to start using chat, GPT and GROK to challenge each other a little bit. All right? Or we're just going to have it refine the ideas. So what we're going to do here is we're going to go ahead. You should look at your business plan by the way. You should read through it and you know, I've got this, this, you know, flow, straight flow, state retreat business plan. Here's my executive summary, right. It's a premium adventure based personal transformation experience designed to help high performing individuals realign their lives, rediscover what truly matters, reignite their internal drive. Blending extreme outdoor adventure with curated coaching experiences. Awesome. 10 to 15 participants per retreat. I've got a vision statement, a mission statement. I've got core outcomes for participants, target audience, unique value propositions. Why is this really unique and why is it really good for me and for, you know, why are they going to choose me over, you know, some other guy instead of going to see Tony Robbins? Right. Why are they going to choose me? Right. What's the daily structure? What are the technology and tools we'll use? What's the revenue model? Marketing strategy. Now I can change all of these things, guys, right? I mean my, you know, according to this revenue model, I can make 100 grand in a week putting this together.

Although it's a ton of work, I've done these kinds of things before. I don't even know if it's worth a hundred grand. It's so much worth putting on Sunday for 15 people.

But it's really fun. They're really fun. So anyway, I've got a marketing strategy, content strategy, lead capture, long term. Like it's, it's built out a business plan for me, right? And I gave it kind of a good length, 1500 words. It's not like crazy, crazy long, but it's long enough that it's pretty descriptive and at least has something filled in for almost everything you would imagine on a business plan. So here's what we're going to do. We're just going to copy that. So right here in the top right of the canvas tool, you just push this copy button. It might fail to clip to copy the first time and I don't know why it does this, but it just does sometimes copy it again and it will work. So then you're going to bounce over to Grok, okay. And what you're going to do here is you're going to start a new chat, make sure you start a New chat. And you're not still trying to build a P5JS platform, brand new chat. Okay, then you're going to type. You're going to type, please evaluate this.

Business plan and tell me where I can improve it.

Okay? Then you're going to use an easy tool where you hold down Shift and you push Return Return or Enter Enter. If you're on a. If you're on Windows, Shift, Return Return. And that allows you to then just kind of take a couple spaces down without sending it to the chat. Then you're going to paste your business plan. Okay? So all I've done, guys, is say, please evaluate this business plan. I'll drop it in the chat.

If you're.

You want to just copy it and tell me where I can improve it.

Okay?

And then, then before you click Go on Grok, I want you to click Deep Search. So it's going to aggressively search the web and it's going to look at all the other retreats out there and see what it thinks, right? It's going to think, see, we're going to see what it thinks about our business plan. So now we just click go and it's now searching the Internet and this thing will come up, says Deep Search Thinking. And it's going to just start figuring things out. So first it's considering. All right, is this a good, you know, let's just look at business plans, right? So it's already looking at, you know, 19 sources on a. Is this a good way to build a business plan?

Right.

How can. Now then in a minute it'll look at, okay, retreats. And I'm just going to do a bunch of research around that and it's going to search, you know, probably, I would guess it's going to search a.

Hundred sources before it's done with this. And then it will come back with some feedback for me.

Okay.

I have a question.

Yeah, go ahead, Brian.

So I just, I had chat GPT a couple, couple months I've been using that. I really like it. And then I just got Grok for the class last night or Saturday. Yeah. And I was trying to compare the two to see what was different.

Right.

And so I was having. Give me the same prompt and then I was having them fight back and forth and I noticed that Gro was sandbagging. So, like, for its statistics, they could say, you know, 100 studies show this, and then, you know, Chad would say something and Gro would go back. Well, 200 studies say the same prompt and kept asking five times. So is, is there a way that I can. I called it out on it too. I said, gro, you're sandbagging your statistics.

You beat Chat GPT, right?

Is there a way to override that? Because I noticed it wasn't pulling all the data, just kind of to sandbag. Is there a way to skip that?

Just explicitly tell it, say I want you to only base this on things that you can find. Right? Don't make anything up. Right? Just tell it right now guys, chatgpt grok these, these AIs are meant, they're built to try to please you and you know, they want to make you happy and so they will sometimes make things up. Right. I have found that for the most part, I would say 95% of the time they're not making things up, they're pulling from real sources. And you can ask it and you can say, you can say this, Brian, you could say, please provide verifiable sources for anything you say. Right? That type of a prompt can be really, really strong to make sure that you're getting really, really good information. Now there are some citations here and things like that.

Let's see here. I want to look at what it gave me here. Business plan strong.

Okay, I'm actually going to tell this, I'm going to say I want you to focus, I'm going to ask again. I want you to focus less on.

The structure of the business plan itself and more and, and, and focus more on the ideas and the market and focus and tell me how I can improve the business plan based on that research.

Okay, so I'm asking it a follow up question. So it give, gave me, it gave me more feedback on the business plan itself as opposed to the actual market. Right. So I, so I asked it again, hey, focus on the market. You might have to do a follow up question on this, but I'm going to ask it a couple questions here guys and then take into consideration.

What it says and we'll show you how to do that really easily.

Hey John, quick question. And I don't know if this is appropriate for this stage of the, of the building part. So I was looking through the Chat GPT business plan and it started pulling financials in here for me. So just started pulling financial projections and things like that. And this is probably coming from a novice entrepreneur, but I'm looking at it and it's like, oh yeah, you're going to be in the whole $363,000 by the end of the year. And so I'm looking at that, and I'm saying, yeah, but hosting doesn't cost that much. I mean, ads and things like that. Customer acquisition cost is probably most of where that's going to go. But that worries me. And I'm going, that's because it doesn't.

Know your whole business plan yet. It's making a lot of assumptions. It's making a huge amount of assumptions. Right now you've only answered seven or eight questions, right? You could update the business plan and say, okay, hey, we're going to. And don't worry about the financial projections. They're not going to be an important.

Part of this exercise.

We'll talk more about that right on day three. We do a lot of sales and marketing and how do we get to money fast and that type of stuff.

But, but I'm telling you, it's not.

Going to take that much. I spent less than a thousand dollars.

Getting to multiple hundreds of thousands with, with, you know, my, my current business.

That I'm working on that I built.

In this way that I'm teaching you guys, okay?

You do not have to go in the whole. It's, it's, it's quite frankly, thinking or it's, it's responding as if we're still in 2020.

Right?

It's not thinking of. It doesn't know the way that you're.

Going to build this yet.

And we haven't told it yet. So that's okay. All right.

Yeah.

Okay. So I just barely. Yeah, I just barely had a good question, by the way. I just barely had it, you know, do some more, you know, work on the, on the market. How can we get better? How can we do better? Now here's what I'm going to do. I am going to copy. Actually, you know what? Before I copy it back over, I'm going to do. We're going to do one other thing, so pay attention really quick. Here on Grok, you've got this deep search thing. You also have think, okay? Deep search is going to be heavy on research. Then think is going to be more analytical. It's like opening 4.5, right? Whoever asked me about that earlier, right? You're, you're turning into a different model and you're going to have it think in a little bit different way and really focus on reasoning. Okay, now think what?

Think about everything you've just presented in your past couple answers and tell me how I can improve the business plan.

All right, so now I'm having to consider the business plan, the market, everything. How can I really improve the business plan.

Okay.

And now it's going to be thinking and thinking and thinking and this might take a minute. I've actually asked it questions, guys, where it took eight minutes so that might happen.

You might get stuck where it's thinking.

For a long time and then it comes up, you know, and you can follow it. You know, you can click expand for details and you can watch how it's thinking and you can try to read.

And follow up with it even though.

It'S like crazy fast. But you can like scroll up and.

And then just kind of read along with it, see what it's doing. Where can we do this? How can we do this better. Market analysis, SWOT analysis, financial projections.

Yeah, see it's focusing a little bit.

Too much on the business plan. Again, I really want it to focus on the ideas. Focus more on the ideas and the market and how we can really stand out.

I'm asking it one more follow up question. So I'm asking it two deep search.

Questions and two deep think questions and having it really focus on, on, you know, how we can make this, make this plan better.

All right.

By the way, this industry is huge.

The wellness retreat industry is 366 billion by 2029. That's pretty, that's pretty big time. So that's awesome.

Okay guys, so Grok, you've just asked it a couple questions, right? Maybe three or four. Maybe a couple follow ups. Maybe on the business plan, maybe on.

A couple things like that, Whatever.

So now what we're going to do is this. We're going to copy the, the responses it gave to us and then we're going to go back to chat GPT and I'm just going to paste it in and then I'm going to again do my shift enter enter thing so I can say something. I'm going to say please refine the.

Business plan based on.

This content. Right? The business plan that we've already made. Based on this content. Okay, Chat paste. Okay, so I just dropped it in the chat and then we're going to do this. So, so now what it's going to do here guys is it's actually going to completely rewrite the canva, right? Or I, maybe it might, maybe, maybe.

It'S going to be right. It, maybe it's just going to make a new one.

No, it looks like it is rewriting it.

There we go.

Okay, so it's going to consider all of the research, everything we just did. Now I, I had to do a couple things right here.

Let's Bounce over to here. Right there. That's good.

Okay, so I had it do some search on the business plan itself.

I had to do some search on.

The, on the market. And you know how we can do better here. And, and so the point is you can take, you can go back and forth between these models.

Right?

Sounds like Brian was already doing and playing with and. And have it challenge itself and then have it refine the business plan and.

Make it stronger and stronger and stronger.

Okay, so after you've done this, now you're going to read through it. If you're actually really building out your business, you're going to read through the whole thing and you're going to say, all right, I want to change this.

And let's clean this up.

And you know, this is missing a piece here. And just talk to A.I. right, like, hey, I really want to.

Clean up and add this piece.

Can we try to refine this? And you know, this is the feel.

That I really want.

I think.

I feel like we're missing it.

Or I hate the tagline. Get rid of the tagline. Or let's come up with a better one.

Or I don't need a tagline.

Whatever it is, just talk to it, tell it, give it your feedback. You're the customer, right? You're the customer. It's going to do what it wants. You know, what you want it to do.

For the most part.

Okay.

It's still considering and refining my plan.

So, guys, once you get done with your business plan and you've gone into grok and then you come back into, you know, chat GPT and you refine it, make sure you read it, make sure it's, you know, there's nothing crazy that it's added in there.

You know, revenue model looking good here. Marketing strategy, street growth plan, key metrics.

Okay, so it's improved it. It's definitely added a few more things here.

I don't love that it's added in a few things here.

Like, for example, it said, hey, let's add in psychedelics and let's add in VR.

And I don't exactly want some of those things.

And so what I can do here, guys, is I can go through and I can figure out, you know, let's.

Get rid of the optional VR stuff. I'm going to get rid of that. Even though I used to have a company in that space.

I'm.

I'm just not going to do that for here.

So you can actually go into the canvas, guys, and you can edit it. Right?

Let's See here, I'm going to search. Oh, I can't search for that. That's too bad. That's okay. I want to make sure. Let's see here. You need. Nope, don't want this. Oh, and get rid of the psychedelics.

I'm not going to do that in my retreats.

If you want to do that, there's lots of options out there. Okay. Return home.

Okay, so you definitely read your plan, guys. Make sure there's nothing crazy in it that you don't like, you don't love.

I'm just gonna get rid of this here.

Okay? Take two minutes. You got two minutes, guys, to finish with your business plan. 10:50, we're done. And you can obviously go through and.

Edit it yourself, guys. But if you find that it, like included something you don't want, you can.

Obviously just say, hey, ChatGPT, take this aspect out. I don't want it.

And I'm actually going to update it so it knows. So it's going to do one more, one more go through on my, on my thing to make sure that I, it integrates.

I said, hey, I hate the psychedelic stuff and I don't want to include.

VR in this to make sure that we're avoiding these things in all of our plans, right?

So it's going to, it's going to give me an updated version here. So, guys, now we have a business plan, all right? And it might, I know you might still be working on it, and we're just going to keep going, but, you.

Know, try to keep up a little bit.

So we're gonna, we're gonna move to another document now. Now we're gonna create a, a dev document, right?

A dev plan.

Now if you just ask it, if you just say, hey, make me a dev plan, right? A development plan, it's going to think that you're thinking like 2020, and hey, I'm going to hire some engineers or.

I'm going to hire some firm to.

Do this for me. I'm going to spend $150,000 and it's going to take nine months. And no, we don't do that. We're building this thing today. It's going to be done today. You guys will all have a business up and running within an hour from now.

And then we'll spend the next two days refining it and building it, making it better and really diving into other plan or other AI tools that can help make it stronger.

But you guys will be ready to rock. So the next thing we're going to Say here, we're going to say I now and I will copy this, this prompt over for you so you don't have to, you don't have to type it out. Okay, so I now want to build.

Or I now want you to build a development plan for me. We will be using Replit to.

Stand up the site.

Extremely quickly. We will use all internal tools to repl it for the database and for auto scaling servers, etc. We expect, yeah, I expect to have this up and running within a few hours from now.

Okay, so I'm going to copy this chat paste.

Okay.

The reason I said a few of these things, guys, is because I want it, I don't want it to make.

A nine month plan.

Right. We're going to do it in ramp.

But we're going to make it automatic. Here we go.

And it's going to be like, oh, that's really ambitious. Right, so now it's going to create a new canvas and I should have said maybe you can even edit into the chat. You can, you can add, you know.

Please write, write this in canvas. Right, you can have it do that as well.

But it's making the dev plan now and going to be built in replit. Replit db replit autoscale enabled, all that stuff.

Great.

Yeah. Question, Travis.

Yeah, so, and maybe this is for like later today or tomorrow. I'm curious your thoughts or research you've done on so replit for like back end and the bones and the meat of it, but then using tools like VO or V0 for front end because I've between the two of them, I've heard V0 creates absolutely fantastic front end in comparison to some of these other tools. Curious your thoughts on that or if that even matters for this course?

It does matter. I think it's a very smart question and probably a little bit above the pay grade of a lot of the people in here, but that's fine. The reality is I have found Replit.

To be extreme, extremely capable of building amazing things.

Can you build some of it in here and then some elsewhere? Yes. But part of the beauty of replit, the reason I love replit, overlovable and.

Cursor and all these other things is.

Because I have found Replit to be.

The most complete solution that can stand.

Up very amazing things. Now can it do the best of the best of the best? Maybe, maybe not. I think it comes down to your prompting. I think people that get bad results in replit are bad at prompting and.

Bad at using the tool.

Because I've seen people complaining. I'm on a lot of Reddit forums with Replit and stuff like that. And I'm actually in touch with Replit, the company and they've interviewed me and.

A bunch of things.

So I've talked to them a lot. And, and the reality is I've seen incredible things built with Replit, like really beautiful websites that are, you know, stunning visually. Right. I mean, my genaipi.org site isn't amazing.

But it's pretty dang good.

Right? Like it's a pretty solid website considering.

What it's building there.

Okay, so great question, Travis Jacob question.

Yeah, so obviously I'm not creating like a site, more of like a product, right. That's integrated with my CRM here at our company. So would I use a similar prompt or is Replit the correct tool for me to be using as well to automate what I'm trying to automate?

Yeah, no, it's definitely. You're building a tool and the tool's web based, so it is a site.

Right?

But just be clear in your prompt, say, hey, I'm building a tool by building this tool, right? It's going to be web based, whatever. Yeah. And then you'll be able to integrate it with APIs and whenever the heck you need to, to be able to.

Connect it over to, you know, Debussly.

Or whatever you're building there. Right.

Or your existing CRMs.

So yeah, just be clear with whatever you're building, guys. It's a website. I'm, I'm copying and pasting my prompts because most of you, you're building a website, building a product, you're building a company, you're building, whatever it is, it's going to build you a full service.

A full scale solution here.

Okay? So, you know, I'm building a website that's going to actually allow people to build and sign up for things and all that stuff. Right? So I'm also going to, I want to, I want to read my, my plan, right? I want to read my dev plan. Make sure, you know, I'm actually going to, I'm going to actually tell it to improve my dev plan a little bit. I'm going to say this, I'm just.

Going to say, please ensure that this website can handle not only the marketing aspect, but also the, let's see, marketing aspect, but also the sales. I'm sorry. Yeah, the sales, the account management, access to, you know, course content pre, pre. During posts and that I can use this to deliver my, my retreats for Everything technical.

All right, so I'm just kind of giving a little bit more instructions and it's going to update the website plan.

A little bit here.

Tell it what you want, right? Say this is the tool I'm going to be using boostly. You know, this is what I'm doing it for me. Be very specific, right? Tell it exactly what you're building, what.

You'Re doing, and it will create an amazing plan in 13 and a half seconds. Right? That's what you're going for here.

Very, very fast iteration fast, testing fast, iteration fast to market. Then you can go in and, you.

Know, be able to graze here. Brian, go ahead.

So mine is. Didn't show up with all the green coding stuff is just words when I.

Totally fine. I actually prefer it to come up with words.

So I wish mine had done what you.

Yours did and whatever, you know, it'll work either way.

Yeah.

If it's not, if it's not green like this, guys, and it still looks like your other plan.

Amazing.

Totally fine.

Totally fine.

Okay, one more thing I want you to do, guys, to your plan. Let's see here. Okay, I want you to do one more prompt. Okay. So again, we'll do this in the chat as well. So please create a. Or sorry, please. Actually, you know what?

We're going to do this.

Please add to the dev plan a feature by feature list of everything that.

Needs to be created, both front end and back end.

Okay, copy. And I'm going to have it update the plan one more time. I just pasted the prompt into the chat so you guys can snag that. So it's making all the features just.

One, one at a time. Well, this is going to be really.

Really important when we get into replit. And by the way, if you do want your, you know, back to the question about, you know, should we use.

This if we want it to be whatever. You know, if you want something that's like really visually stunning or different or.

Better or whatever, use that in your prompt. Say, I really want to focus on the visual aspect.

Like this should be an experience when.

You'Re on my site.

I want it to be awesome. Maybe all of you should say that.

I mean, maybe all of you want.

It to be awesome, right?

If you're building a tool internally, maybe it doesn't matter as much.

It just needs to be really functional. But if you're building something for Evan, for his. For his business in Hawaii. Right? Yeah.

Let's make this really visually stunning and exciting and get people you Know, ready.

To, ready to sign up, right? Inspiring.

Okay. All right, guys, take another two minutes, finish your, your dev plan, maybe refine it a little bit. Look at it, read it. You don't need to understand a lot of the stuff that it's out. You know, it's talking about databases and back ends and the scaling and you don't have to understand any of that. What I really just want you to look over is, does it feel like, yeah, this feels about right. Like, here's the plan for the landing page, here's the plan for whatever, right? So you should have at this point a business plan and a website plan.

Yeah. This is awesome. Oops. Yeah, I'm feeling really good about this, guys. This is looking nice. All right, as soon as it hits. Oh, there it is.

I was gonna say as soon as.

It hits 11 o', clock, I'm moving forward.

It's 11 o', clock, so. Hey, we're already two hours in, guys. We're flying, we're flying, we're flying, we're flying. Now we get to you. Now we get to jump into replit rep. What's gonna be really fun. Now let me just say this. If I was building out my actual business and I really, really wanted to spend time nailing this, like I spent a long time talking to the different chats about just my business name and trying to nail that. For Gen aipi, I had the General AI Proficiency Institute. I wanted to make sure that my business felt, you know, really legit, I guess, I don't know, was that when I was building that out? Right? The, the institute, the word institute obviously made that really nice and using a dot org made that feel really nice. Even though I'm. It's not a non profit, it is a for profit company. It is something, you know, I'm building, training for companies and individuals and all these different things. I wanted to make it a dot org, right? And so just having that branding and feel also led to a lot of design decisions that I make later on about, you know, how do we make this feel really legitimate.

Right.

I want to be able to assess and train and certify and put out research and things like that. And I, that all revolves around my brand, right. And what I'm doing here. And so if you're building for your own brand, you have to consider those types of things. Another thing that I didn't, that I didn't go over, that I, that I would have done during the branding exercise is I would have bounced over here to.

Well, here, let Me go over to this one right here.

I would have said, all right, let's open Facebook. And I need to look and see if there's something called Flow State Retreats, right? There might be, like, that's a pretty. That's a pretty solid, you know, Flow State Retreats. Boom. There is a page that already has this. They have zero followers, so I'm not going to worry about it. And I'm going to call my. Think Flow State Retreats. And they only have six. 1-5566. This is their name, right? They haven't actually built this out. Someone had an idea that wanted to do this. This dude is competing with me. That's my competition, guys. I got to take him down. Okay? So now at least I have something I need to push over. Also, I noticed he's in the UK now. I can come here. I can look at a few more. There's a Flow.

There's another.

Okay, here's retreats in Australia. So, hey, it looks like the UK and the Australia maybe have something there, but I need to do that. And then I'll come in and I'll just say, all right, let's build a new page, right? And I'll click on, you know, wherever it is that I click on to do that.

See more wherever.

And I build out the page, right? I go to pages here and I'd say, create a new one. And I would start doing that. I would check. By the way, I jump into Instagram as well. I want to bounce in here and I'm going to search for. Let's see here.

Where's this search for Flow.

Whoops. Not glow.

Flow State.

No, we're going to go.

Yeah, we'll type this. Flow State.

We'll just see what happens with Flow State. Flow State leadership. That's me. That's awesome. Flow State Studios. We'll type retreats. So in my. In my branding, right, in my branding exercise, I'm probably going to add this in. I'm going to go and I'm going to look for, you know, the socials and things like that. If I'm trying to do this, I'm also going to bounce over. And by the way, there is a Flow State retreat, so I'm going to click on it. I just want to know. I want to know what's going on here. Yeah, this is that same guy. 130 followers. Snow, surf, skate. All right. Mindfulness retreats. UK, Portugal, France. Right? I bet this dude's cool. I bet he's. I bet he's got Some fun stuff there, right? So anyway, maybe I'll even go to this thing, be like, hey, we're both Flow State retreats. Let's do something together. And then I'm gonna bounce over to godaddy. Okay. I'm bounce over to GoDaddy.com. i'm going to check the brand, right? Is flow state retreats retreats.com available?

Flow state Retreats. Yeah, I think I typed that right.

Okay, it's taken. All right, that's interesting. Let's go to the website. Flow State retreats.com. okay. Looks like it's for sale for 995 from somebody. So I'd have to figure out who owns that. And if I really love this name and I really want that, I can do that. Now the reality is, guys, for me, I can just drop it on my flowstateleadership.com and say retreats. And I just have a tab, a section on the site, right. That might point at what I'm building here. Or maybe it's just a link and it points over to another site. So I can do this in a lot of ways. Now let's say I came up with another name that was like, you know, flowy. Let's say I wanted to call my thing Flowy. That's the name of my app. Okay, so I look for Flowy. I bet you flowy.com is taken. Okay. Floyd world's available. That's fine. Flowy.com is taken. So I can't do that. But what if I go, you know, flowy. Let's say I'm also, you know, I change it from Retreats. And it's actually, you know, it's an app about, you know, mindfulness and things, whatever. Maybe I'm trying to make something like that. What if I just call it Flowy app? Hey, it might be available. Floyd. Flowyapp AI is available. Flowyapp IO is available. There's a lot of different things here. Okay, I can also say getflowy.com.

I think that's funny sounding.

Okay, that's also taken. I could say Flowy co dot com. I can keep just playing around with this until I find exactly what I want. Obviously someone likes the word flowy and they've tried to go through and get a bunch of things here. But you guys can take your business ideas, check the socials, check the website if you're actually doing that. And then if you're actually building a website, well, bounce over to, you know, just say, hey, you know, file an LLC in wherever you are right. And then skip the legal zoom and all these things here and bounce over to corporations.utah.gov and follow the instructions. I know it might sound or feel difficult, but it's actually crazy easy to.

Set up an LLC if you've never done it.

We're not going to go over that in this class, but it's super easy. You'll set up an LLC and then you'll say, I need an ein. Okay? And an EIN is your. Is your employee identification number that you get from the government. It's free. It's super easy to set up. An LLC is about 50 bucks in some states. It's a little bit more some states. In Utah, it's about 50 bucks. Okay? And the only reason you're going to need an EIN or an llc, guys, is if you need a bank account, right? If you need a bank account for your business, you need to set up, for example, Stripe. If you want to accept payments from people, you're going to need these things. You'll need an ein, a bank account, all that stuff. And that might take a little bit of time. Now you can do all of this stuff. By the way, guys, I did all of this stuff in three days. When I did my. When I did my challenge, I started Thursday morning, and by Saturday it was done. I had an llc, I had my ein, I had Stripe set up. I had a bank account. I had all these things, and I did it in between all of my prompts instead of sitting there and watching Grok think, I type and I bounce over. And I'm working on something. I'm working on setting up social, I'm working on an image, I'm working on whatever. We haven't gotten into logos yet. We'll do that tomorrow. But I just wanted to say, really at a really high level, right? There's a few business things to set up. And by the way, if you have questions on this and you really want some guidance on it, please tell me. And I can get you some good information from me for my course that I have on this. Along those lines, guys, we have a. There's a folder here. Build a business in a weekend with AI.

I've got a bunch of stuff I'm.

Going to give you guys that you'll have at the end. And so if you feel like I'm not getting all of this right, and I want to be able to go back and check on something, not only will you have the recording, but you'll have these guides and all these different things, software and tools and overviews and all these fun things that will help.

You get through this again to be able to.

To. You know, I've got a few. I've got one really advanced workflow that's.

In there that's pretty cool that I just added in there.

So. And I'll keep updating that folder even over time. Time. Okay.

I don't know why this is frozen here. I was trying to start a new page. I'm not going to actually start it, so I won't worry about it.

All right, so we are in chat. GPT. Let me bounce over back to here. We've got a business plan and we've got a dev plan. Okay, so what we're going to do is this. We're going to actually. You know what, we got to do one more thing. We do one more thing. We got to say this, say. All right, we'll close out of the business plan here and we're going to say, please write me a three sentence.

Description of this, of this, this website and business.

Okay. All right. Three sentence description. Just ask it for a three sentence description of what we're doing.

Right.

Flow State Retreats is a high end adventure retreat platform that helps high performing individuals reconnect with what matters most through.

Immersive experiences in nature.

Blending intense outdoor challenge with personal coaching, AI tools and post retreat integration, it delivers lasting clarity and life realignment. The website serves as both the marketing hub and operational backbone, handling applications, payments, course access and participant progress from start to finish. Okay, that is amazing, guys. I've just built a whole new thing around Flow State Retreats which has been on my mind for a while. I wanted to do some official retreats through Flow States. I actually have a good description of.

What I'm trying to build here.

So I've got three sentences now. We're going to bounce into replit guys for yes. Travis, you're getting ahead of yourself a little bit. So what we're going to do, guys, is who here is not signed up for replit yet? Is anybody not signed up yet? If you're not, I'm not. I'm not seeing anybody's hands raised.

At least I didn't think I did. Yeah.

Okay. You need to be signed up for it. Sign up for the core plan. I think it's 35 bucks a month. If you use my link that I provided in that, then you get $10 of free credits and so do I. Thank you. I promise you I'm not getting rich off of those credits.

But you know, it helps a Little bit.

Okay, so we're bouncing into here and I'm going to open a new file. So when you're in replit, you're going to be sitting right here. It's going to say, hey John, hey Jessica, hey Eric. What do you want to make? Okay, describe an app. Right here. You're going to paste your three sentences. Okay, don't click start building yet. You're going to paste those three sentences. Then we're going to bounce back to chat GPT and we're going to grab my business plan here. I'm going to copy it. It's going to fail to copy the first time. For whatever reason, I'm going to copy it. Then I'm going to bounce back over here and I'm going to paste it and it's going to paste as an attachment like this. Okay, then I'm going to come back over here and I'm going to close out of that.

Oh, man. Dang, dang zoom controls. Man, I hate these things.

Always in my way. Okay, then I'm going to come to my website plan. I'm going to copy it and I'm going to come back over here and I'm going to paste it. Now I have a three sentence description and two pasted plans. I have a business plan and then I have a development plan. Okay, then what I'm going to do with this, guys, is I'm going to go shift, enter, enter here and I'm going to say, and I'll write, I'll copy this into the chat for you. Please build a or please, sorry, please.

Yikes.

Please follow all of the guidelines provided.

In the attached text also.

Actually, yeah, well, right now that's all we're going to do. We will do a readme and some progress. We'll do some MD files in a minute. But right now we want to get it going. Okay, so we've attached two of these things. We're going to follow the guidelines and the attached text. All right, Copy. So this is. So go ahead and do it, guys. Go ahead and dive in. All right, I'll, I'll copy this little prompt into the chat. Super easy. Please follow the guidelines in the attached text. So copy your business plan, copy your dev plan and drop it into replit and then click start building. Okay, I'm gonna do it right now. You guys can watch on screen. You guys are probably doing in your own already. Okay, and your three, your three sentences, if you're building an internal tool, right, say that, hey, please give me a three sentence Description of this tool that.

I'm building for Boostly or whatever it is you're doing. Okay.

Okay, so it's, it's analyzing the provided documents to understand what we're looking to build. It's going to create a plan for.

My Flow State Retreats platform. It's going to be thinking for a moment.

All right, Then it says build the initial prototype, right? Here's the plan. Now before you guys click Approve plan and start, if you're to this stage, okay, it's going to ask you if you want any additional features. Hey, PostgreSQL database. Yes, I want that. AI journaling tools. Maybe. Maybe later on. Actually, I'll say yes because I want.

To integrate some AI into this so.

I can show you how to do that. Wearable health tracker integration. I'm not going to do that. Advanced analytics for retreat impact measurement. Sounds fun. Why not? Payment plans and subscription options. I'm not going to worry about that right now. I'll let Stripe handle that.

I'll let Stripe handle my payment stuff, so I'm not even going to worry about that.

So I've got my build the initial prototype and then a few options I'm going to click. By the way, if you don't love something or you want to add one more guideline, you can message it before you approve the plan and you can tell it to give you another plan. Right? Most times going to be okay. Now guys, I want to make sure if you don't have this postgress SQL database integration thing here, maybe tell it that, maybe, maybe message the agent and say, hey, I want postgres SQL database integration. Now, those of you building internal tools, maybe you need that, maybe you don't. But for the purposes of this class, I want you to build it out because it's very, very powerful to learn.

How to do this inside of replit, to be able to set up your own database and do all these things.

So make sure to include what I just pasted into the chat.

PostgreSQL database integration for long term data persistence. Okay, we'll talk about that means later.

Right now I just need to make sure it's in your plan. Then you click Approve plan and start. Okay, so this is where the magic really starts to happen, guys. It's going to build a website and it might be good, it might be awesome, it might be bad, might be ugly, I don't know. But it is now coding and starting.

To build hundreds of things, thousands of things, thousands of lines of code to.

Get this up and working, it's going to focus first on. On some visual stuff. It's just going to kind of do a wireframe outline. It might be good, it might be nice, it might not be. Don't worry about it. Like, you can see, if you're looking at my screen, you can see that my menu, my. My top menu, bad spacing, whatever, right? But hey, it's got a picture of a mountain. Strip away the noise, step into the real you. That was the tagline that we came.

Up with, that I came up with.

With chat GPT, right? Premium adventure base, whatever. Apply for the next retreat. And it's just building this website like right in front of me here.

Okay.

Really, really amazing. It's thinking through. It's kind of trying to do its best, you know, attempt at building what we've told it to build. Now we've given it some pretty dang good instructions where people fail with replit. And if you go into the Reddit, you know, discussions on replit, they're saying, yeah, I tried it and it sucked.

Right?

Well, guess what? Their prompt was make me a weather app Go. Right? I mean, that's all they do. And then it builds something that's okay. It doesn't work. Nothing's happening. There's no direction. Right. It's just like it's going to do random things. By the way, if all nine of us on this call here, all typed in, build a weather app, it would build nine different things in nine different ways. It would follow nine different ideas, and it's just not going to be that great. All right, how many people have. Have pushed go. Have pushed approved plan and start? Looks like pretty much every who is. Not yet, Jessica. Not yet, Evan.

Not yet.

Okay, where are you guys stuck? I want you guys to get to this spot and then we'll step away.

Go ahead, Evan.

I was trying to find my business plan again to copy it over from chat gdp.

Do you find it? You just have to scroll up back into it.

It should be just.

I just told it to send me my. Or create my business plan again or just send me. Okay, so that's.

Just writing it out again.

Okay, that's fine. Perfect. We'll give you. We'll give you a second to catch up. Okay, we'll. We'll talk here for a minute. I think this is a good spot to talk. Jessica, how are you coming? You have it up and are you. Are you in replit and pasting stuff in?

Okay, very good, very good.

Okay, guys. Pretty amazing, right? Isn't this Awesome. I mean, I just think it's so, so cool how fast it can just start to prototype something for you, right? Even if you just have an idea inside of your business or you're in a design meeting, right? Like for your business or whatever it is, you're with your team and you're like, hey, I wonder if we can build a tool really quick that does this. I don't know. Let me pull up replit really quick. Here's what I want to do. You don't have to go through the whole chat, GPT and all this stuff. You can give it good instructions if you just start typing into it, telling it exactly what you want, right? Like, hey, we need a calculator that like thinks through this, this and this and creates a model for this. And it's like, okay, give me a second. And you know, in the middle of that meeting, four or five minutes later, you've got something working in your meeting that you wanted to prototype there. All right, what was the other thing we're pasting? Okay, yeah. So, Brian, you're pasting the business plan and the development plan or the website plan?

Development plan.

Okay. Yeah, yeah.

So then you're asking it, you're asking it to, you know, fall. You know, please follow all the guidelines provided in the attached text.

And then you paste that three sentence.

Description of what you're building and you.

Tell it to go, okay.

I'm not a huge fan of the, the color scheme that it chose for me. It shows kind of a, like a forest kind of idea here.

It's not bad.

It's not bad. Like, it's, it's just fine. Especially for a look, there's John Cheney, the founder. So you know, I'm black. So you might need to change a few things that it does. And you will.

Absolutely.

That's what we're going to work on later. Right? But, but it gets this stuff, it gets this stuff going. I, I could even tell repl. I could say, hey, please search the Internet, find a real picture of John.

Cheney and replace it.

Right? And I could give it a few more instructions. I can also drag a picture in.

And say, hey, use this picture for John Cheney instead of, you know, instead of the one that you've used already.

So we'll do that stuff later. But I probably will tell it. I'm going to say, you know what, let's switch this to kind of more.

Of a blue, blue colors.

And by the way, I don't think in this time sometimes I usually do, but usually in my Business plan or my dev plan. I actually go over a little bit related to colors and I don't think I gave it any colors. And so because of that. Okay, well, I can't expect it to know what I'm thinking. That's my fault. I did a bad prompt, I did an incomplete prompt.

We'll say, right?

I should have said, hey, I want you to follow a blue scheme and a water scheme because flow, flow state, right? It's water, right? And so I'll be able to tell.

It that in a minute.

And I'll say, hey, we need to.

Switch some things out.

I want you to follow a better, a better version. I might even have it start completely over into a whole new project and.

Just paste my prompts in again. I'm actually going to say this. Please update the dev plan and the, yeah, you just the dev plan to include a bluish dollar scheme so that we can roll with the theme of flow like water.

Okay. All right, so it's going to update.

The dev plan a little bit here and then I'll be able to, if.

I want to, I can just have.

It build a whole new thing, which I might, I might do because it's really not hard to, to make that happen.

Okay guys, we're cranking, we're cranking. Hopefully what you've learned so far in the first two hours of this class is that you can use multiple AIs.

To work in concert together.

You're the conductor, right? I'm a musician, right. I've got a piano right here, I've got a piano right there. And I've got one sitting up against my wall over there by my Lord of the Rings.

Sword.

I'm a musician, I write music. And last year I had the opportunity to go see John Williams live in concert.

Thank you to my, my brother in law on here and my sister who got that for me.

But my wife and I went out there and I got to see him.

Which has been a, been a lifelong.

Dream of mine and it was amazing. So there was another conductor that conducted for kind of, you know, the first little bit of the concert and then.

John Williams came out.

But you know, those musicians played well for the other conductor, don't get me wrong, world class Philadelphia Orchestra, really amazing. But when John Williams stood up on that thing and he, he just picked up his thing, like you could just see the musicians respond a little bit differently. Like they just respected him more because he's such a good conductor and composer.

And obviously he's done so much for.

The world when it comes to music and movies. And so that's what your role is here, guys. You were the conductor. You're in charge. If it's doing something bad, it's because you told it something bad, right? You got to. You got to work on all kinds of stuff. Look at this.

It made a little bouncing arrow for me here. So cool.

Okay, now I want to talk a little bit about agents. Sorry, guys, I'm coughing a little bit here. I want to talk about how these agents think and how they actually. How this AI, like what this AI is actually doing here. And I'm going to turn off my.

I'm going to stop sharing for a second here so I can just look at your beautiful faces while we, while we do this.

Maybe Jessica's the only one that can.

Take that description, but, Travis, you're beautiful too.

So, guys, let's. Let's pick on someone here. We're going to pick on Devin. Devin. We're pick on Devin. All right, Devin, I want you to tell us your get.

How do you.

How do you get ready in the morning? Just tell us your routine.

How do you get ready?

Oh, my get ready routine is awful. I sleep in until my kid, my 5 year old comes in, says, dad, what about breakfast? And I take 10 minutes to get out of bed, grab the toddler, make breakfast, stumble out the door in a heap.

Fair enough. Do you, do you shower ever? Yeah, yeah, we shower. Or brush your teeth. I'm sure you do a couple of those things, right? So you skipped over. You don't really think about it. You don't really think about, you know, even the things that I mentioned, right? Eating breakfast, maybe showering, brushing your teeth. You just know to do that because you're a human. You've got a smart brain and you've done it so many times, you've done it thousands, tens of thousands of times even. And so you're an expert. You don't really think about it. But now let's zoom in on one of those things. Tell us how you brush your teeth. Try to be as specific as possible.

I brushed my teeth with my left hand because I'm right handed, so I just want to, like, increase my dexterity. So I brush my teeth with my left hand. And so I brush my teeth with my left hand. I use a, like, just a typical, you know, battery driven toothbrush from Sonic Care. Brush my teeth. I start at the top, I go to the bottom, I brush my tongue, I spit everything out. I do it one more time, hit with Mouthwash and I'm done.

Okay, perfect. Thank you. So what you're doing there, Devin, is without thinking about it, obviously you're doing something called chain of thought reasoning. Okay. And that's what this AI, this replit.

Is doing, chain of thought reasoning.

And what that means is it takes a task that might seem, you know, a high level task, for example, getting ready in the morning and then it breaks it down into a bunch of tasks. Well, my five year old wakes me up or my alarm, I hit snooze two times and I set my alarm early so that I can do that or whatever your routine is, right? I get out of bed and I say my prayers, I read my scriptures or I meditate for a minute. I walk out and I get a drink of something, I have breakfast, I brush my teeth, I do my mouthwash, I shower, I get ready, I get dressed, right? And then obviously with each of those things there's sub subtasks, right? Like the brushing the teeth, right? You say brush teeth, okay, we're going to turn the water on, I'm going to get my toothbrush wet, I'm going to, I'm going to open the cap, I'm going to put my toothpaste on, I'm going to close the cap, I'm going to put it back, right? Each of those little things is a little task that our brain or our body is doing, right? And it's. So we've got tasks, subtasks, sub subtasks and, and we create all these different tasks and do all these different things. Then we do it naturally because we have supercomputers in our brain or in our head, right? Our brain is a supercomputer that can do that very, very easily. It can do very, very advanced chain of thought reasoning as opposed to just a simple response. Most people on here, I would venture to guess right now I know a couple of you are a little bit more advanced. Most people use AI today like, like a glorified Google, like hey, I asked Google, I asked you to everything now and that's great. You're already ahead of people that still aren't using it, right? There's a lot of people that haven't even done that, but a lot of people just like, yep, one quick question, maybe a follow up, okay, we're done, right? But when you start to learn about all these other tools and then you start to get introduced to tools like replit or manus things that we'll play with some other really neat tools here that do very advanced chain of thought reasoning. It can, it can open up a whole new range of opportunities for you. So, so back to repl it, right? Relet is thinking about a high level task, right? And it's, you know, it's not working right now because it's working on some back end stuff and trying to get things working right. And so it's going to, it's going to work through, it's going to troubleshoot it, but it has this high level task of hey, build, build this thing right? And now we gave it some good instructions, some good guidelines. At least now I didn't guide it as, as well as I wanted to because I didn't tell it a color scheme right. So I might go back and just.

Say, hey, rebuild this again with a different color scheme.

And, but then it's thinking through and I want you to look at what it's doing. Here it is. Let me, let me see if I.

Can click on this. Open, open, open. Edited, edited.

Yeah, maybe it doesn't even have, maybe.

It doesn't even have all the things that it's doing yet. It's building some storage stuff, index is doing a few things here, but in.

A minute it's going to start building out so many different parts of this.

App and this website that it's just.

Building everything it knows, it thinks, it knows, it says. Okay, so for a website I'm going to need a nav menu and I'm going to need a login button and I'm going to need a hero image and I'm going to need a main message and it's finding some errors and it's, it's working on stuff. It's going to have some errors as.

It gets set up guys, just let.

It, let it work through the errors.

Okay?

But it's going to have, you know, the main message is going to have the subheading, it's going to have all these different things until it's, until it's completed what it will call, you know, your first mvp. And then we'll start working on, we'll start working on how we actually make it work and make it refined. But, but guys, understanding that chain of thought reasoning, understanding what it's doing helps you be able to better guide also can help you stop it from doing stupid things as you get better at it. If you're reading and you're like why are you doing, I've literally stopped it and said why are you doing this? We don't need to do that. Like it'll spend 20 minutes and some of my credits building out a huge test environment, you know, to test like one feature. And I'm like, dude, we don't need to do that. Just tell me where to push and I'll push it. I'll test it for you right now. If you want to automate testing, great. You can set up a ton of unit tests or pen tests or whatever you want to do. We'll talk about. Those are later. But. But you're in charge. That's what I'm trying to say. You're the conductor. If it's doing something wrong or not. Like you want it.

Well, did you tell it well?

No. Well, then don't get mad at it. Right? Tell it. Be a good conductor. Be a good prompt.

Engineer is what it was, is what it's really called.

Okay, does anybody have a cool website that's popping up now they want to share? Like just the first draft.

It wants me to integrate with open API.

You what? You what, Devin?

It wants me to integrate with open API. Like, I need an API key.

Okay, I will come back to you, Devin. We'll do that in a second. Travis, share your screen really quick and then Evan will do you. And then Devin, I'm going to come to you and we'll work on that. Okay, we'll work on that problem.

Okay, let's see here.

Here we go.

Okay, can everyone see the screen?

Yep.

Okay, I'll take you through the top just super, super briefly. So I'm going to log out because I want to see. So you can see my landing page. So this is the social media tool I was telling you about. And then it wants me to log in. And this is crazy. And it just stores some fake user login information. And this is just super bare bones. I'm still kind of testing to see. Well, is it exactly what I want it to be before I launch? It's got a search feature. So say, okay, I want to find an influencer to collab with Butic Cosmetics. And then you put in your customer avatar. And then I want this to be in Georgia because I live in Georgia. And so I want to target local people in Georgia and get some sales there. And then it asks, how big do you want this person to be? Do you want them to be a hundred thousand followers? Three to thirty. I actually want small followers because my dollars can go farther and there's higher trust in that audience. So that's where I'm going to stay. And I want stories and reviews and oh, yeah, let's throw a YouTuber in there. And then you find the match and then it'll think. And it says, here's your person. This is what it's going to cost to collaborate with them. Just roughly, roughly.

Will you look really quick and go to Instagram and look up Garden Guru.

And see if that's a real person?

I'm.

This is. This is not. This is just test data.

Okay.

All right, all right.

I do have. I do have an API installed on the back end where it will return real people.

Oh, nice.

And so, but I. Since I'm not launched yet, it's not pulling actual people. This is just test data.

Very good.

But it proves the concept and says it gives you two things to do. Now. I can contact this person and save them. I'm going to save them, of course. So there's my saves page, and now I'm going to contact them. And this is really cool. I need to clean this up, but it'll generate a template that sounds like a human, where all I have to do is put in my product type, put in my demo, say, help you find this message. Well, and then it kind of hits the top end. I don't know that I like how long that message is. So there's some kinks to work out. I did a little research and found out it would be super convenient for users to now actually message through this. But we get into some regulatory legal issues, and I don't want to deal with those at this stage. So the way around it is literally just a copy and paste, copy, and then the user pastes. And we've navigated all the way around all those kind of regulatory issues, and boom, just like that in seconds. A small business owner now has someone that can reach out to you. Sees the engagement rate. So sees, like, are the. Like, are the people that they're talking to actually engaging with with this person. And social media, around 3% is considered good.

Right?

Again, this is just dummy data. So just threw in 6% for us. You know, gives you a couple metrics. And then like, that they found someone and they have a way to reach out to someone that they can just copy and paste and boom, there we are.

I love it.

Very, very cool. Yeah, thanks for sharing, Travis. I'll stop your sharing here. Is anybody else, like, super amazed that.

He built something, like, in like a few minutes? I think it's crazy.

I'm amazed. Evan, go for it.

Well, mine, I don't know what happened. It was there. Now it says we couldn't reach the app.

Yeah. Yeah. So it's Going through. Yeah, there we go.

There you go.

There you go.

Yeah. As it works through stages, it's going to do that occasionally. Right. It's going to be restarting the app and turning it off and building things and making it more complex.

Let's see if mine is even. Yeah, mine is not waiting, but that's good. All right, let's see what you got.

I got one. So of course that's. Yeah, this made up a cool logo there. Not really cool, but we'll fix.

We'll worry about images later.

That's great.

But anyways, yeah, it just kind of came in here. Aloha. Waichu's Ohana Tours. Right? Local expert guys, comfort and safety, unique destinations, super cool fonts, to be honest with you. And then you can, like, book these tours. Of course, I'm going to have to put in, like, my stripe links and everything else and figure out how to do that. Well, that's. That's a pretty big price tag, so. But pretty cool. I'm assuming it'll be cool to get in here and figure out how to just change each one of these images. Right. And. And kind of get what you want. But I like that it's laid it out in a really nice way. I just need to go in and be able to change the content myself or have it change the content to wherever I want it to be. So.

Yep. No, this is really, really awesome. Yeah. And you can see there's a lot of things to scroll at the very bottom. Here's, here's. Here's one weakness that sometimes replit has. It looks like there's a bunch of text here, but I bet it's a bunch of white text. Try to select text in that whole bottom section. Just like highlight it. Yeah, see that? So, yeah, we're gonna have to fix that. Right? Obviously it's stupid to have a white bottom and then a, you know, white text in the footer. So anyway, very cool, Evan. Thank you for sharing. Right, so guys, we're just sharing some first. First drafts here. We're gonna, we're gonna just stop at that. But, Devin, I want you to share your screen now.

Hey, John, real quick, how do you know when it's done building your site? Because I think, I guess agent.

Yeah, the agent in the bottom left is going to be like, hey, what do you think? You know, what should I do next? You know, and. And so you could just start talking to it. You can follow its instructions or you can say, nope, I want you to work on this. You can say hey, the footer. The. You know, the text is white and the background is white. Please fix this. We'll worry about that in a minute. Okay, I'll start to. Don't work too much on your website, guys. I mean, you can if you really want to get ahead, but I'm going to show you some great ways to.

Do that in a. In a nice way.

So with the website, like credits, does that take up like 10 credits, or do I have to, like, constantly upload credits or whatnot?

No. Yeah. So you guys paid, I think, $35 for your monthly whatever, right? And. And then you got $10 of free credits if you used my link. If you didn't, then whatever, you're going to pay for those credits. But they'll just come out of every month. You get $25 of credits. When I built my website, it took me about $87, my entire company, and I'm talking 30 hours of development in repl. It. It cost me about 87 bucks in total. Okay, so you guys have all probably spent somewhere around 25 cents so far of your credit. Okay, so, yeah, find a developer that can work for that. Cheap Evan. All right, Devin, go ahead and share your screen. We'll walk you through this. OpenAI integration. Okay, guys, we usually teach APIs on day two, but we're going to do this really quickly. It looks like you've already opened up a tab up there with Overview OpenAI API. Okay. That you already logged in. This is awesome. So go ahead and scroll down a little bit here.

Where's the API key?

There should be an API key. Maybe it's start building.

And that top, that green button at the top? Maybe. Maybe click there.

I swear they changed this thing. Okay.

Yo, you haven't done the settings, have you?

Have you done this part yet, Devin?

Yeah, I just did it. It's in the settings.

Okay. Yeah, I think he still has to do this. I think you still have to create an organization. So go ahead and name your organization whatever you want, and then you know what best describes you not technical? As a financer.

That's great. It's okay if you say that.

That's what I click.

That's what I say.

All right, well, invite my team later and then, yeah, call it whatever you. I mean, that's great. Generate. API key.

Awesome. Yeah.

Perfect.

Generate.

Okay, go ahead and copy that API key right there and then bounce back into replit and paste the value. Click continue and then click add to account secrets.

Bottom left there.

Boom. Now it's got your API Key. So if you ever build anything else with AI and you want to use that API key, it's there. Okay, let it just think for a second. Bounce back to OpenAI really quick. Let it think. It's going to have a couple errors. Go ahead and scroll down here. Click next. Whatever.

I know that things. The zoom thing is in the way.

Drag that out of the way there. Yeah, yeah, it's just hiding there. Okay, so we're going to stop sharing your screen now. Add five credits. When you use open AI, you got to add a credit card in so that it can actually, you know, do that. So I'll stop sharing your screen so you can go ahead and type in credit card stuff. We don't need to do that anymore. But that, that's one important thing. Guys, whenever you set up an API, the amount of API usage you're going to use during development, even if it needs you to put a credit card in, is probably going to be a grand total of like $0.11 or $0.03 or maybe a dollar if you're using something crazy expensive and doing a huge amount of requests. APIs are usually very, very, very cheap. And they're built to be able to scale to thousands, hundreds of thousands, millions, without breaking your bank.

Okay?

And so most requests are going to be a fraction of a fraction of a penny to be able to do something. But you do have to put a credit card in. Like if you're using something like Twilio because you want to send automated text messages, or send Grid because you want to send emails or Stripe or whatever. Stripe actually doesn't even require a credit.

Card because you're just accepting payments. It's a free one.

So there's free APIs, but most of them are pretty dang cheap. I actually did build a Weather app, a really nice one, in one of the other classes. And I had to get a couple APIs from like weather API.com and weathermaps.com to get some visualizations, things like that. I spent like 7 cents in the.

Entire development thing that I was working on.

Right? So they're very, very, very cheap. But the APIs allow you to connect to other data and other services in a very powerful way. We'll talk about APIs more tomorrow. Okay, sweet. Now go back to Replit and watch it do its thing. Right. So we're still back on the replit again. We'll do eight. We'll do API more API stuff tomorrow. Okay, I'm going to share my screen here, guys, and we're going to go Over a few just, you know, tips and tricks here that will kind of get you going. If you want to keep working on this, you know, throughout the day, you're going to be welcome to do that. There is a document. Sorry, it's going to be over here. Software coding. Quick. Start really quickly. We're going to do a fast, very fast coding boot camp, okay? You guys are going to understand exactly what's going on kind of, right? And you'll have this document. I'll, I'll copy it. In fact, let's do it right now actually.

Let's see, we'll click the share button here. Man, these things are always in the way. All right, copy link and I'm going.

To share this with you guys so.

That you have it so you can.

Look at it later. Go ahead and copy that out. I will share the link later on, but for now, copy it. Open it on your, on your own thing. Okay? So the reason I do this little boot camp, guys is not to teach you how to be full on developers. It's, it's so that you can give replit faster and better instructions. Okay? So if you say, for example, hey, will you fix the button over here and the way that the screen looks isn't great, or, you know, whatever, or you could just say, hey, I don't like the way the UI looks on this element. Can you fix that? Right? Or do this to it, make it red, right? Or hey, the UI of this section is just bad. Please redesign that, right? That's what you're going to be doing now. You're going to be acting as your customer, right? Going through replit, going through your app, going through your experience and just considering, hey, what can I do to make this better? All right? And so being able to use some of these words will help you do that. So front end again is everything that you can see, right? Everything that you interact with, right? Whether it, you know, forms, clicks, navigation, typing, whatever. A few of the technologies you'll see that is HTML, CSS, JavaScript, React. You don't need to know what these are, okay? But just know that if you see those, at some point, maybe you might come across this. If you don't know what something is in replit, ask it, ask it, hey, what do I do here? A lot of you guys are going to start asking questions, John, what do I do here? What should I do on this? And I'm going to say the same amount of effort that it took for you to ask me. You could have asked replit, right? And it's going to give you a better answer than I could. And you don't have to wait for me, you can just ask it, right? Hey, what do I do here? What do I do on this? How do we make this better? Right, so, all right, so, front end, like a login form where users enter the email and password, right? You've got, you got, you've got two little blanks and you've got a login button, right? And maybe a forgot password button. Those are all front end elements. Then the back end is going to have other languages. Python, Node JS Express.

You don't need to worry about these.

We are using a PostgreSQL database, right? And then Firebase is stuff that Google uses. There's, there's lots of ways to, to build these things.

There's thousands of tools and there's no.

Way we could cover them all. But what you need to know is the back end is behind the scenes that processes data and handles requests from the front end. So when you push log in, after you submit your email and password, it's going to send a request to the back end and say, hey, does this combination exist? No, sorry, wrong password. Or this combination doesn't exist.

Yes.

Hey, great. Here's your authentication key, here's your token. We'll let you in, right? So it gives a little signal to the system and says, let them to the next step. Let them inside the backend. Communicates with databases, processes data, manages business logic, et cetera. Then you have databases. We asked it to set up what's called the PostgreSQL database. You don't even know what that is, other than the fact that it is structured data that looks like an Excel document. It's a table. You've got columns and rows and you've got data. And that data will determine how your app works. Right now. You don't have to think through this. I'm telling you guys. It might sound hard right now, but after a couple days here and working on it and building your own data, or actually having REPLIT build your database, but you're managing it, and being the director, you're gonna be like, I know what I'm doing. I know how database databases work. Now, will you be capable of writing your own database? No, not yet. But will you be capable of looking at what your. What REPLIT is building and saying, I know what that's doing.

Yes, you will be able to.

Okay, there's also. You can query databases, you can ask IT questions. You don't have to know how to do this. Know that replit can for you. You can ask it a question and.

It will answer things for you.

Okay. One thing that you do want to. So APIs, we'll talk about this really quick and then. Yeah, I'll get back to the other thing. I was, I was getting ahead of myself a little bit. APIs are, in very simple layman's terms, a connector. Okay. They connect one thing to another. They're an application program, integrated, who cares, right?

They connect things.

So just like I was talking about with my weather app, I needed weather data. Well, I don't have weather sensors out there. I don't want to put satellites into the air. That's too much work. Okay. So I'm going to reach out to someone else who has done that and who is pulling that data and then feeding it. Right? And they'll do what's called an API and then they'll charge me for it. And that's great. For every request, they'll charge me a fraction of a penny for something. And so I can then connect to those APIs and have those APIs pull into my app. So now I can have real time weather data inside of my app. Maybe that's important for me. Evan, you might want to grab a weather API and have something that shows the weather for where you're doing your services. That could be a real world use case where you could pull an API and add a little bit better, make your site stand out a little bit more. Hey, this guy's really thought through it. I even know what the weather's going.

To be for this week.

Right? Or you can have it pull it into. Hey, you're reserving for this. The weather is usually according to historical data. This. Right. You could have AI analyze that historical data and then present it to the, to the user and email them, you know, their, their confirmation with, you know, expected weather, even if they're, even if they're booking five months out, right? I mean, you could do all kinds.

Of cool stuff like that to have.

It automatically handle all of that. Now an API, guys, is a URL where you're pointing at. So just like you saw Devin set up the OpenAI thing, that actually wasn't a URL or an endpoint, that was actually a key. It's pointing at, you know, I don't know, chat.OpenAI.com or whatever it is, and then it's giving it a key, a unique string of numbers and letters that's going to be, you know, your password, basically.

To get in.

Okay, APIs grab data and they grab data, they write data, they update and they really all come down to this, right? Get post, put or delete or crud, right? Create what? Create, remove, update, delete. No, create.

I'm messing one of them up.

Someone fixed me.

That's, that's a developer. My, the R is escaping me right.

Now, but it means the same stuff. Okay, to get you, first you get data, very simple. You're retrieving something, hey, what's the weather in Hawaii? I'm grabbing the data, right? And then I'm going to display it.

On the front end, okay.

Or maybe I'm going to post it, right? Maybe, maybe I'm going to create a user, right? So you type in a user or whatever and okay, the R is read. Thank you, thank you, thank you. And then the. Anyway, so the post, you're going to create a user. So you create the user, you create a password. It's going to post or send new data to the database. Here's a new line of data. And now we have user ID number two.

And we've got this user ID and.

We'Ve got this password, et cetera. Right, then put. Maybe you forgot your password. So we're going to update it. We're going to go through this, we're going to authenticate a different way and we're going to send an email confirmation and then get a link that is a put functionality. It's going to update that password from whatever it was to new data and then delete. We're going to remove the data. So you're either grabbing data, sending data, updating it, or removing data. That's basically like the entire Internet. That's all it does, right? It shows things, pulls things, updates things and removes things. Okay, Authentication, you know, you don't need. You can, you have access to this, you can read it later. The only thing I'll say here is you have your standard kind of token based authentication and then you have OAuth, right, an OAuth. So your standard one is, you know, username, password.

Oauth is login with Google, login with Facebook, login with whatever.

Okay, AI agents, we don't need to do hosting and deployment. We are going to talk about this a little bit. Hosting and deployment is awesome. Replit can handle everything.

That's all you need to know.

Okay, all right, that's all we're going to do on that. Guys, I know you might have some questions and I know you might not understand all of that perfectly, but now when you're talking to Replit, when you're updating things, when you're working in there, you're going to understand just a little bit more about how to talk to it. So I'm actually going to bounce back into Replit for a minute here. It looks like it has, you know, fixed this now. I could tell it right now to, you know, update it to blue and water things. I'm not going to do that for the moment because what I want to look at really quick is I want to see if it's set up a database for me. So I'm going to click on this little new tab thing. I want everyone to do this as soon as you. After you watch me do it. Database. Okay. It is not set up a database for me yet. A lot of you are going to get to this point here and it's going to say, hey, create a database. So I'm just going to click Create. Create a database. Okay. Now it's. Now there's a database set up, but there's zero data in it. But we're now connected to a new database that's being served. For those of you that are interested. From the Neon database, you know, there's a. There's a company called, you know, if you just go, neon Database.

Neon Tech is what it is.

Yeah. So there's, you know, this is the database. It's easy. If you have a question about how good is the database? It's awesome, right? It's really good. Neon's a really, really good one. But that's what they integrate natively into here. And then you're going to say, I just activated the database. The post. Oh, yeah, the postgres SQL database here in Replit.

Let's make sure the app is working with the database.

Okay. I will send this over to you guys in the chat so that you can say that. All right, so go in, turn on a database just like I did, especially who was sharing before the. Was it Travis? Yeah. Make sure you get. You have a database set up. Now, it might have already set up a database for you. I asked it to set one up in my plan and it hasn't done it yet. So that's okay. But go ahead and do that. Go ahead and get your database going. All right. It's not going to be that hard, but it is going to have to redo some.

Some ideas here. Okay.

All right, guys, we've gone through a heck of a lot in day one. We have one more thing we're going to do. In replit before we're finished here. But I'm going to let it do the database first. The database is critical. The reason the database is critical, guys, is because if you don't set up a database, it will build everything in your app to feel and look like it's working, but it won't actually be working. All right? It's going to put everything in what's called MEM storage or memory storage. And guess what happens to memory storage the second you hit refresh?

Refresh.

It's gone, right? And so you'll create a username and a password, and you'll create all these cool things, and then you'll click refresh or it'll reload and you'll be like, where's all my data? Where's my. Why is my username and password not working?

Well, there's no bit database, right?

What it's focusing on in the beginning, like I said, is the front end. It's going to make it look like it's a real thing. Then you set up a database, and then we'll start pointing things at the database. But again, in order to get things to point at the database, all you have to do is copy my prompt. I just activated the database. Let's make sure this is working with the database.

Awesome.

Boom. And now it'll set everything up. It will build out all the tables it needs. And you're going to build more tables later. It will build more tables later. You're going to think through it. You're going to build more features. But the reason why you want something to work with a database, guys, as opposed to MEM storage, is because you.

Want it to remember data persistence, right?

You want it to persist or stay in between sessions, right? In between turning your computer off, turning back on, refreshing a browser, closing a browser, sending a new link to somebody, right? You know, whatever you're trying to do, opening it on a new computer, you want it to work everywhere. And so what happens is a database stores all that data, and then no matter where you're coming from, it's looking at that database, and it can then log in, log in, and everything, all of the data will persist. And so, Evan, you're saying, how do I update my prices? I don't want to charge $14,000 for this. Well, the database will store all of your pricing, right? You need to make sure that your. Your vacations, your tour options, your tour times, your everything are all going to eventually be in the database. And a nice, you know, not crazy complex, but some simple, relatively complex relationships that, that will, you know, give it logic on how it builds everything out. Right. So you'll have a user that signs up for a time, or actually you'll really look at like bookings, right? Let's say you have a table for bookings, right? And on that table it will have user ID time, you know, product id, you know, which might be product three. And product three on that table is, you know, volcano tour, right? And then here are the times available, right? So it'll have all this data. You don't have to think through that. You'll want to understand it. And I'll help you look at it later so that we can all. We'll take time with everybody just kind of looking through things. You'll want to look at the database.

And figure out how it's building things properly.

So right now what it's building is an auth provider, right?

Or my login.

Right, my login. And it's making sure it's using the database now. Okay.

I don't even know if I have data.

Yeah, I've got some data now. Applications, journals, modules, retreats, user modules.

Users, right?

I don't have any users yet. I have no data in any of these. But I can go and I can start looking at it, right? Here's the, here's the retreat, here's the module. Let me look at the retreat. Is it have a price? Yeah, it's got a price right there, right. It's got a capacity, it's got a description. Right. And so again, Evan, back to you. If I'm updating my tour, my Ohana tour, well, you can update the description right here, you know, or you can tell it to, you know, do it on the front end. You say, hey, update, update this description for this product to this. And then, and then Replit will do that for you. It will open up the database, find the right spot, update it, and then say done. And it takes seconds to do stuff like that, right? So you can tell it to do it or you can understand, hey, I can click on this little plus button and I can manually go in and.

Look at things if I want to get crazy.

We'll work on, we'll, we'll, we'll click on other tabs and other things in here, guys, but, but, you know, to start here, this is what I want to do now. The last thing we're going to do today, guys, and I want you all to pay attention. This is critical. Someone asked earlier about a Readme md. This is important. This is another reason why when you go into replit and you just say, build me a weather app, it doesn't work very well, right? It's because it gets lost. And so in order to use any of these AI tools, these complex agents that are doing, you know, maybe multiple hours of work, you. You start something and you leave. There's, you know, if you use Manus, Manus can do hours of work. If you use deep search on ChatGPT, hours of work. So if it's going to do hours of work for you while you're, you know, playing around a golf, you need to make sure you give it a really good prompt so that it's not wasting all that time. Right? So here's how we're going to do it. Now, if you look on my screen here, guys, you'll see. So you guys see this agent in this preview, there's a little button right here that says open files in the top left.

Okay?

In these files there's a bunch of just random stuff that you don't need to understand at all. Okay? But you'll see an attached assets thing and you'll see these pasted things, right? And if I click on it, hey, here's my executive summary. And hey, here's my dev plan. Amazing. Okay, so what I'm going to do is I'm going to click back onto the agent here and I'm going to say this. I'm going to say, Please create a README.M. i'm going to copy and paste this over. Okay, Please create a README.md file. With the attached business plan in the attached underscore assets folder. Also create a progress.md file with the.

Attached dev plan in the attached or.

No, in the same folder. We'll say in the same folder, okay, copy. And I'm going to say go. And then I'm going to send it to you guys here in the chat. Okay, you guys all pasted those things into the attached assets. It's going to, even if you didn't call it dev plan or you didn't call it business plan, whatever, it's going to look at the documents, it's going to analyze which ones which, and it's.

To going to make it okay.

So now it's going to create a readme MD file and a progress MD file. An MD file, guys, is just a type of file that's used in development to help give guidelines, usually to developers, to people, right. That will read that so they can familiarize themselves with the project, the agent or AIs operate at least they communicate with us in the same way that people do with words and with instructions and with guidelines. Right. So in the same way that you would give guidance to your development team, you're giving guidance to this, to this, to this AI agent that's doing complex chain of thought, reasoning and thinking and doing all kinds of crazy things for you. Okay, so we will let it do this readme file and then this progress file. Now if I click over here, actually I'll wait for a second. I'm impatient. I want AI to be faster. It's already, you know, I'm already annoyed that it takes me, you know, hours to create a business.

Even though it used to take months and hundreds of thousands of dollars to, you know, build something up like this.

Okay, it's done now. It's done for me. Go ahead and look at my screen here, guys. Unless you're still typing, but readme md Now I got this nice, beautiful thing. Flow, state, retreat. Strip away the noise. Step into the real you executive summary. All the stuff that I worked on and built, it's there. Here's my progress MD file. Here are all the things. Okay, now there's just a bunch of check marks on all of this stuff. Yeah, here's the next step. Yeah, so it's showing me everything that it's done. Postgres integration, drizzle ORM database migrations. Okay.

Yeah, it's pretty good. This is pretty awesome. I'm actually going to look though at.

The.

This to make sure it's doing this well. Yeah, it's doing, it's doing pretty solid here.

Okay, so basically guys, this, these two documents are going to keep replit in line now. Now it's going to keep it in line by talking to it and explicitly.

Telling it to listen to those.

So you can start a new chat at any time. When you start a new chat here in replit like this, it knows nothing about what you've previously talked about. It's like you're bringing a brand new developer onto the scene. Okay, that's important because that means you got to give it some instructions. So at the end, by the way, at the end of a chat, before you go over to a new chat, I'm going to refresh. It's going to pull up my old chat here. What you want to do is you want to say please. Now we just did it, so we don't need to do this. But let's assume you worked for like an hour talking back and forth to it, telling it to do this and change this and move this button and make this work. And hey, I clicked on this and it didn't work.

And et cetera, et cetera, et cetera.

Right, so you're going to say, please update the progress MD file with everything we've done and be sure it reflects.

The current state of the project.

Okay, Copy and paste. Paste. Okay, so that's in the chat now if you guys want to copy that somewhere, wherever you're keeping your notes for this class. Now, a lot of these prompts are in some of the documents I'm going to give you, but not all of them, right? Not all of them. So copy and paste over what you want before we end today. I'll give you guys a couple minutes to do that right now. But, but then, you know, it'll update everything. It'll get it kind of all up to date. But now when I get into a new chat, this is the prompt you're going to give it. You're going to say, please read the progress MD and the readme md, whoops, read me files and familiarize yourself with the project. Now let's work on X, you know, whatever you want to work on. Okay, so I'll copy this prompt over as well, guys, and drop it into.

The chat so you can take notes. Drop those into notes on your computer.

Or a Word Docker, Google Doc, or whatever you do.

Right, Notion.

So, so now I'm going to say, now let's work on, you know, whatever I'm going to do. Now I'm going to show you guys a couple more little tools here and then we might go like, like five minutes over here, guys, I want to, I want to just cover one more little thing. Okay, so apply for the next retreat. Learn more. All these things say, I want to, I want to work on the menu. I don't like the menu. I think it's kind of terrible. I don't like the spacing. And so instead of saying hey, now this is a pretty easy thing, you can say, hey, the nav menu spacing is bad. But there's a really neat tool here in Replit. If you look in the, in the agent chat, if you see my screen right here called the element selector, the element selector allows you to just tap that thing and it turns blue. And then I can put my mouse over anything I want, right? And say, hey, this is what I want to work on. Okay, so I want to work on the header, right? The whole header menu is just not great. Okay, so now let's work on making.

The spacing for the menu, for the header menu better.

And then what I'm going to do to be even more clear is I'm going to take a screenshot on Mac guys that shift command four and then you. It brings up this little pointer and then you can select, you know, a little piece, grab it, and then drag it into this chat right here and it can see what's going on.

Okay.

The spacing of the words and buttons is not great.

Please fix this.

Okay. And you know, I'm just going to send. Get it to work, right? And it's going to start working on it. Okay, perfect. I'm. I'm on it. Okay, so that little element selector, as well as using screenshots, I don't always select elements and I don't always screenshot it. It depends on what I'm telling it to do. But if I want it to update like some specific text. Absolutely. I'm going to click on this element selector, I'm going to say this text. I want it to be a different font or, you know, whatever. Okay. So you can tell it to do whatever you want and be very, very clear. All right, so now it says, I can see from Both the Progress MD and ReadMe files that this is a premium adventure retreat platform with multiple features already implemented. Great. So now it's like, great. Now I know what's going on. Right? It's immediately up to date. It knows what's happening, and it is.

Going to build amazing things for you.

Okay. All right, guys, to finish up today, I, I want you to. You can spend the. Really, what we wanted to get through today was this point we're building in repl it. We understand how to talk to it a little bit. You guys are probably going to struggle today. If you try to use it some more, you might not struggle. Some of you are gonna be like, this is amazing.

Right?

And some of you be like, I'm.

Not really sure what to do here.

And I got stuck on this thing. Fine. I want you to bring your struggles to the class tomorrow. We'll talk about it, do some screen sharing, do some debugging. A lot of tomorrow is going to be playing with how do we get more advanced? We're going to go over image generation tomorrow, logo generation. We're going to get, you know, how do we pull those things from those other platforms we'll use to build those.

Things, how do we pull them into this platform, etc. Etc.

Okay, so the last thing I want you guys to think about is and, and, and we'll just kind of end today with this and we'll start it with a bit of a discussion tomorrow. I want you to really think about, you know, your experience in especially those few that are entrepreneurs and hired developers before and been through this. Like now all of a sudden anybody.

And everybody can do anything, right?

And I'm not saying like a Tony Robbins, like anyone. You could do anything, you know, like you can build anything. Now any human being, my 10 year old can build anything he or she can imagine, right? She can, she can get on and build a game sitting there during church. She's done it. She built a little temple matching game sitting in church, right? And it took her five minutes to do it, right? And you know, of course, you know, those of us that are adults and are a little bit more developed, may have business experience, whatever, we can build anything we can imagine, a fully functioning, you know, piece of software. And we're going to teach you guys how to make it really, really awesome over the next, over the next couple days. But think about what that does to society, right? You know, what is going to happen to developers and bankers and lawyers and all these different things, right? As agents get smarter and smarter and smarter and smarter.

What is that going to do? What is that going to do to our society? What's going to happen as this, as this rolls out?

We'll talk about that more tomorrow guys. And I don't want you to answer the question right now. I just want you to think about it. Think about how even just your ideas around AI have changed in the last three hours and we'll talk about that.

More tomorrow as well.

For today. Well done. I will, I will send a recording of this to the email, everybody's email and so you'll be able to, you know, Jacob, you can catch up on.

A couple things you missed in the.

Beginning if you want. You, you pretty much got it all.

But, but just kind of some, some high level things and then yeah, if.

You guys have any questions, feel free to respond to my email. Many of you have my phone number, text me. In fact, you should all. I think it's in my email that I sent out last night. But yeah, if you get a question like hey, I'm stuck on this, what do I do here? Like won't bother me, I'm running to the doctor right now to like follow up on my arm that's hurt and whatever and I have a massage later this afternoon so I'm not going to answer during that.

But anyway, other than those things, guys.

Like you said we could just ask Replit.

Ask Replit. Ask Rock. You guys are experts now. Go ahead, Travis.

Quick, quick one on asking Replit. I've done this before. If you ask it. Hey. Hey, did you do this thing, or. I'm not sure what to do from here. Sometimes it'll just automatically do. So you have to be specific and say, I do not want you to code. I just want to know what. What happened, or whatever. Like, I've. I've had to say, hey, did you set up the database? And then it's like, I didn't bump. And then it started doing it immediately.

Setting up the database.

Yeah. Yeah.

Which was helpful because I wanted it to. But if you just want to know what's going on.

No. So draft is pulling from my replic, you know, tips and tricks document. That's in. That you're going to get tomorrow. But. But, yeah. Don't write any. You know, here's my question. Don't write any code.

Just tell me.

Right? Don't run any code. Just tell me if you want to ask it a question. Like, hey, I want to fix this. What's your plan to fix this? Let me approve it before you do it. Don't write any code. Just tell me. Okay, great. And then it'll write some stuff for you, and it'll say, here's my plan. And you could say, great, go. Right. Or you can say, ed, modify this, then get started. Right? So, yeah, otherwise, Replit is overeager. It wants to work. It's gonna. It's gonna get going. Right.

Great point, Travis.

Any other questions before we finish up, Guys? If you feel overwhelmed, just go, relax. Go on a walk right now.

Go look at some grass, and you'll.

Feel a little bit better. And then come back to it and. And text me. Text me. Tell me. You're like, hey, I'm a little bit behind here, and I'll try to help you out. Okay. All right, guys, thanks. We'll see you tomorrow morning, John.

Thanks, John.

Appreciate it.

Thank you.